After impatiently waiting to get accepted as a beta tester for ChatGPTs latest API features (higher token limits in particular), I decided to set myself a challenge to see how much of the flow of running a YouTube channel could be automated. A task I set myself over the course of around 10 days was to create an Autonomous and faceless true crime youtube channel. The only input for a new video upload should be the name of a serial killer, E.G. "Ted Bundy". The output should be a full 15-20 minute video which is consistent to the brand. The channel should output 1 video per day, with the goal of ad monetisation within 3 to 6 months. All aspects of the creation of the content should be autonomous, such as: <ul><li>Research into the investigation</li><li>Creation of the transcript in the correct format</li><li>Synthesising speech as audio from the transcript</li><li>Using Ffmpeg for all media encoding, concatenating, conversions, audio &amp; visual FX, complex filtering and overlay tasks.</li><li>Transcribing the final audio track back to textual transcript with timestamps for every spoken word</li><li>Inference of sentences within the transcript to extract keywords for context</li><li>Categorisation of a stock footage library by keyword for keyword matching appropriate videos to overlay at certain contextual timestamps</li><li>Creation of AI images for overlays, thumbnails and a channel avatar</li><li>Creation of the talking avatar video from the AI generated still image</li><li>Creation of metadata (uses my previous AWS Lambda Youtube Metadata Generator Project code but natively from within the node app)</li><li>Utilising Google APIs for automated YouTube upload</li></ul> Each stage within the process saves data to file, so that each output can be inspected and each stage can be re-run in isolation. There are further plans for a feedback loop whereby i intend to setup a basic nextJS site to host a simple poll which will house a database collection of proposed crimes/serial killers to do videos for. The YouTube transcript and description will include links to the site so that users can vote and also submit new entires for which serial killers they would like me to do a video on next. The idea being that this can be an automated process and the video can be generated and scheduled for release within 15 minutes (but in reality a content release strategy will be set). I will not delve too deeply into the intricacies as to exactly how I have managed to create this channel due to the fact it is an ongoing business venture, but feel free to ask me in person or during interview and I will be more than happy to demonstrate and elaborate.

As the developer of the YouTube Meta Generator Chrome extension, I created both the frontend Chrome extension and backend serverless AWS application designed to automatically generate metadata like titles, descriptions, and meta keywords for YouTube videos and video files. The application uses AWS Lambda Functions &amp; Layers, DynamoDB, S3, SNS, and OpenAI APIs (Whisper ASR and ChatGPT) to provide backend services for the client-side Chrome extension. The main features of this project include a frontend Chrome extension built with React &amp; TypeScript, automatic metadata generation for YouTube videos, transcription of video files using OpenAI Whisper ASR, metadata inference using OpenAI ChatGPT Completions, video processing using Ffmpeg and YTDL, storage of video files and metadata in an S3 bucket, and a serverless architecture using AWS Lambda Functions &amp; Layers. This allows for the automatic conversion, transcription, and inference of YouTube videos and video files simply by uploading a local file or pasting a YouTube link. I set up several AWS Lambda functions, an S3 bucket, a DynamoDB table, and an SNS topic to manage the processing and storage of video files and metadata. The application relies on OpenAI Whisper ASR for transcribing the audio content from video files and OpenAI ChatGPT Completions API for generating metadata based on the transcribed content. Deployment of the application is made easy through an AWS SAM template, which provisions all necessary resources, permissions, and IAM roles for the Lambda functions. To deploy the application, simply follow the standard AWS SAM CLI deployment steps, providing the necessary API keys and credentials for OpenAI APIs. I may at a later date implement a paywall for a paid version of the extension via StripeJS. This paid version would allow bulk upload and usage for YouTubers that require automation at scale. Currently I have architected the system to developer specifications to ensure I do not incur high fees (limit the possibly amount of function invocations and DB read/writes), but if this goes into production, as it is a serverless system it will be very easy to scale it up. The extension is currently available but not publicly listed. I plan on releasing it publicly once I can increase the length of the video to be analyzed to 1 hour (currently there is a limit of 15 minutes due to upstream bottlenecks and limits with the OpenAI API's but higher limits are already in beta. You can try out the chrome extension at this link: https://chrome.google.com/webstore/detail/video-metadata-generator/kefolmegbkljjphcfefocncgohjhgdhh This was a fairly short project which i did in my spare time over a few weeks, predominantly just as a bit of fun to enhance my skills in Typescript, AWS serverless and chrome extension development.

I created this portfolio project using my favourite tech stack, consisting of React, Redux Toolkit, NextJS (currently at version 13), MongoDB, Mongoose &amp; Material UI. This project is designed with Material UI in Figma, fully written in Typescript, tested with Cypress and also has documented API routes with Postman &amp; OpenApi3.1 The site is composed of the frontend pages, the API which interacts with the MongoDB cluster, and the backend CMS for managing my projects and skills. This project utilises Redux (toolkit) for state management as well as React context and hooks.I have implemented some Amazon Web Services into this project to handle simple tasks such as email sending (SES), and to offload longer computations which would otherwise hit Vercel's 10s serverless function invocation timeout. This site also has a custom Cron schedule using Amazon EventBridge &amp; Lambda to allow me to reach out to the GitLab API once daily to build my own custom GitLab activity monitor table (as seen on the homepage). Site speed and performance was optimised using the React profiler and carrying out performance optimisations with NextJS, such as; static site generation, ISR (incremental static regeneration), code bundling, optimised request caching, image &amp; video optimisations, lazy loading and dynamic imports of scripts and components. Tests were carried out (in production) using tools such as lighthouse and GT Metrix and I was very happy to achieve extremely performant results across the board.

This is a personal project I did as a proposal to Boomtown Festival to try and help them solve some of their sustainability issues, namely the problem of many of their festival goers leaving behind their rubbish which is a huge expense to clean up as well as damaging the environment. I planned to solve this using a system that allows users to prove their campsite area is clean by uploading an image of it which would be verified using AI. In return for a clean campsite, the user can get extra perks such as secret stage access and cheaper tickets for the following year. This project was built with React, Redux Toolkit &amp; NextJS, Minimal UI for frontend components (based on Material UI) and integrates with a MongoDb and Mongoose backend. other features it uses are: Google Maps integration, QR code creation &amp; read for ticket scanning and pulling QR codes embedded in images as well as Exif data extraction from mobile devices to check when an image was taken if it is a screenshot and to extract embedded geolocation coordinates on it for plotting on a map of the festival site. teachable machine, (an extremely fast way of integrating TensorFlow models for demo purposes) trained and used to classify and detect litter in campsites. This project is purely a proposal for the boomtown team to view, so it will not end up on a production server.

Picks &amp; Parlays consists of a very large legacy Drupal website with high traffic (an average of 1M monthly visitors). The traffic fluctuates a lot depending on factors such as the sports seasons, day of the week and time of day. The site has a very high throughput of posts from multiple content creators (anywhere from 20 to 70 per day approx) I was part of a 2 man dev team on this project, with my main responsibility being the all of the backend including:<ul><li>The migration itself</li><li>Database migration from Drupal to MongoDB &amp; continued management of the MongoDB cluster</li><li>building the backend CMS</li><li>building the backend API</li><li>building a data import pipeline to continually import &amp; posts from the legacy site to the new one (because content editors will not move to the new CMS platform straight away).</li><li>building and maintaining a range of external API services for fetching sports data and integrating it within our database</li></ul> As well as the backend, I also worked on some areas of the frontend leading up to and after the launch to ensure everything went smoothly and any issues were quickly addressed. The infra &amp; tech stack we used on this project is:<ul><li>A custom CMS. Initially this was in React, Redux, NextJS &amp; Material UI and API</li><li>A Sports data API (initially in firebase, which i rebuilt in expressJS then migrated to the NextJS API folder</li><li>A frontend repo built with Vue, Nuxt &amp; Vuex &amp; deployed to AWS ECS &amp; Fargate</li><li>A data pipeline utilising AWS Lambda</li><li>a sitemap generation cron job using AWS Lambda &amp; S3</li></ul> Some important notes on this project were:<ul><li>Legacy posts had be pruned and migrated due to SEO value.</li><li>There was also a lot of other important SEO areas to think about such as proper redirection of migrated urls.</li><li>Prior to joining the project, the system used Firebase, which was causing a lot of problems for us working at the enterprise level due to limitations pre-indexing complex queries and how firebase handles iterable data, so I proposed we move to a better database solution such as MongoDB, then I carried out the work</li></ul> I implemented E2E testing with Cypress for major CMS tasks as well as API endpoint testing. All endpoints are well documented using Postman &amp; OpenAPI specification.

This was a quick project using React, NextJs &amp; TypeScript. The frontend CSS framework used was tailwind CSS.This site has been built in a simple SPA style and is not currently public as it is awaiting client feedback but can be accessed on the staging domain.

The Magic Mirror is an open-source 'Internet of Things' project. I built the magic mirror using a Raspberry Pi as a web server in Electron. I developed this project locally in GitLab, then SSHed remotely onto the device over wireless and set up a simple CI pipeline whereby I git pull the repository &amp; any new changes onto the device. I run this project using Pm2 so that the server automatically restarts itself if the device is powered off at the mains or restarts for any reason. The device itself is just a simple raspberry pi with some extra ancillaries such as a small monitor screen, a full 21" monitor for the back of the actual monitor, and some speakers. The mirror itself is an electron instance. It comprises various built-in and custom built modules such as an alarm clock, calendar, daily &amp; weekly weather forecast, TikTok profile stats, Fortnite account stats, daily images of earth from space, and Spotify Connect integration. The Spotify connect integration connects to a Spotify account and allows two-way remote control access of the account so that the mirror displays what is playing and can also be used to change the music. My time limit on this project was tight, so I did not implement any voice controllers or speech-to-command type system, however, if I had more time, my next big feature would be to add something like Amazon Alexa or similar to allow easier control of things such as Spotify playlist selection, specific track search, setting alarms, adding dates to the calendar and putting the device in standby mode.

I created this project whilst conducting research &amp; development to demonstrate the benefits of new NextJS features to the rest of the development team, such as static site generation (SSG) and Incremental Static Regeneration (ISR) at build time. I also used client-side data fetching on the front end to offload JS &amp; components with no SEO value from the main thread so they are only loaded as and when they are in the viewport. Along with implementing other NextJS features, this showed great improvement in tightening performance metrics.

Audiolancer.io is a dedicated music industry marketplace that connects professionals and streamlines the purchasing and licensing of audio assets and services. The platform, set to launch publicly in Q3 of 2023, empowers artists to generate income directly from their audience through various means. As the founder and sole developer, I have used my skills as a developer and solutions architect to plan, build, test and iterate over this platform, building it entirely in my spare time. The platform includes a CMS, API, and frontend, and supports sales of vocals, samples, services, and jobs. It is built using MongoDB, Mongoose, React, NextJS, and Redux for state management, and employs a variety of technologies for enterprise-level performance. Payments are handled through Stripe Connect, which enables users to sign up as vendors and use the site as a marketplace. The platform's payment system is designed to handle fund holding, fee splitting, cancellations, refunds, and subscriptions. Assets are served via AWS S3, with Elastic Transcoder and Lambda functions ensuring fast, safe, and secure music streaming. Signed e-PDF agreements are generated using Lambda functions and Node Canvas libraries. The messaging system uses WebSockets for real-time communication and incorporates SMS, email, in-app, and web push notifications based on user preferences. Algolia search is employed for efficient content indexing and synchronization with the database. Authentication is implemented with NextAuth, replacing Amazon AWS Cognito for a more streamlined experience. The platform generates audio data, such as waveform information, BPM, and musical keys, using npm packages and external APIs. WavesurferJS displays waveforms and audio players on the frontend, with added complexity due to their interactions with modals and React's lifecycle. The current CSS framework, ReactStrap, is being replaced with Material UI, and the platform is being upgraded to TypeScript. Integration with external APIs like Spotify and SoundCloud provides additional features, including widgets and "like-gates" for free sample distribution in exchange for Spotify followers. Audiolancer.io is rigorously tested using Cypress JS and Jest, with a CI pipeline for continuous integration and deployment to Vercel.

Tinder Hijack is a personal portfolio project I developed to reverse engineer the Tinder dating apps REST API as well as to look at the app from an analytical perspective. I formed a plan to build a wrapper around Tinder that leverages the API to get extra usage and features out of the app that a normal user of the UI would not have access to, such as automation and greater exposure to other users without paying for the privilege. Most of all, I wanted to see if it was possible to administer Tinder programmatically in a server environment without the use of any kind of UI whilst creating an interesting portfolio piece that almost everyone can relate to at some point in their lives. Furthermore, this project was a great way to showcase newly acquired skills in machine learning with TensorFlowJS as well puppeteer, canvas, face-API, python scripting, teachable-machine, and other various node-based packages for data and image manipulation. The first part of the application was to build a classifier.This was used to sample 10'000 swipes, download the profile data, and store the images and swipe results (which way was swiped) for each image. Together this data essentially formed someone's particular swiping characteristics and romantic Taste. I cleaned up the data by by creating a programmatic data cleanup pipeline that:<ul><li>ensured the classifier knew which person was the subject of the image in cases where there were multiple people in a single image. (using node sharp)</li><li>removed any images using Snapchat filters or similar</li><li>auto-cropped images to show only the subject</li><li>changed images to monotonal to aid the training of the model by reducing unnecessary image data</li></ul> I then trained the model using a simple true or false (swipe left or right) system with the result being a classifier that can receive a set of input images and make a fairly accurate prediction as to what I would do if swiping manually. The system was set up to give a percentage value of the likelihood of a left swipe or right swipe as well, which was useful for further development whereby I set a range of 40 - 60% which was a "not sure" zone to try and accommodate for any anomalies. Once the classifier was created and I had a model I could now build the API itself.I used NodeJS and ExpressJS to build endpoints and essentially built a wrapper around Tinders REST API. I exposed various functions such as:<ul><li>scrape data</li><li>scrape Insta handles in data</li><li>scrape images</li><li>geolocate (known as passporting in Tinder)</li><li>swipe, and Message</li></ul> The scrapers are fairly self-explanatory. The geolocation was a great feature as it really worked with Tinder's algorithm at the time to increase matches without actually swiping. For example, Just leaving the app to geolocate into every major city in the UK, every hour without doing any actual swipes could produce a lot of matches purely just by showing my profile within other people's decks as someone new in the area which was quite an interesting revelation. The 'auto-swipe' feature was initially set to swipe randomly until the classifier was built, I then swapped out the 'swipe randomly' algorithm for the new model trained with images from the classifier based on my personal swiping history and the accuracy felt fairly reliable. The dataset of 10k profiles sounds like a lot (and it was to classify them) but in the reality of dataset sizes for machine learning, it is actually quite a small dataset for what is actually a fairly complex task. Swipe results and any relevant data got saved to a MongoDB cluster so I could monitor activity and prevent accidentally messaging the same person multiple times. The message system simply checked the Tinder data to see if there were any new matches to send a message to or if there were any outstanding replies that needed to be replied to. For new messages, I created an array of boilerplate chat-up lines that the API could send to the user. the plan with this was to tally which chat-up lines got the most responses and save this data into the database just out of pure curiosity to see what kind of messages were more likely to get responses. A further step I wanted to incorporate into this but was a bit out of the scope of the project would be to use a more complex classifier that would recognize objects or subject features to allow the chat-up lines used to be more contextual and precise. At this point, I ran out of time with working on this project as it was purely a fun side project I did during my time off. However, the next stage of the project aside from tightening up the machine learning was to add the ability for the bot to reply to messages. currently, the bot can only send out chat-up lines but it has no intelligent way of replying to messages with any form of context or flow. My plan was to integrate either an Aws or Google chatbot and a Machine Learning layer to continue with flirty conversation without being exposed as a bot. The data extracted from the user's profile, images &amp; previous messages could be fed into the model's parameters to help it make more informed decisions on its replies. From doing some research I found a few examples of systems that would be great for this kind of task such as OpenAI's GPT-3 or Chat GPT which has previously been used for similar experiments. The results of this project were pretty impressive. Not only was it a great learning experience to code, but I also see a lot of potential to monetize, this kind of system, mostly due to the fact the systems metrics (in terms of likes and matches) were about 10x better than the output from Tinder's own paid features such as Tinder Gold, Super Likes and Boosts. This code repository is not something I will be making publicly available due to how easily it can be abused and how quickly it would get me into all sorts of legal trouble with Tinder, however I can explain it and show elements of the repo upon request for any potential hiring managers. More to the point I think dating apps like this are best left to humans, but regardless of this, I have decided to mention the project here as I am quite proud of the work I put into it. If I were to develop this further the aim would be to make it easier for anyone to train a model on their swipe history without having to do any technical work. Ideally, the model would cumulatively become more accurate the more the user Swiped through with some form of feedback loop which constantly analyzed any new manual swipes.

I worked as part of a 2 man team to overhaul the existing casinoguide.se site. We upgraded NextJS to the latest version and updated the site across 3 different repositories (one for each domain). This work mainly involved creating new front-end components such as toplists.

This project consists of around a year of work on various trading algorithms &amp; EAs (trading bots). I initially started working with MQL4 (MetaTrader Query Language 4) for the MetaTrader 4 application before I moved across to MQL5 because of its greater potential to work with many other assets aside from forex, such as commodities and cryptocurrencies. I developed, visualized, back-tested, forward-tested, and optimized a wide range of EAs. I used my own EC2 servers on AWS to run these bots and recorded all the results using MyFxBook. The majority of my trading systems were fully algorithmic, however, I did also create a few indicators &amp; systems designed to give signals for discretionary traders to make the final decision on. I learned a lot about developing trading algorithms, including:<ul><li>General Discretionary Forex &amp; Crypto Trading</li><li>MQL4</li><li>MQL5</li><li>working with time-series data</li><li>Creating strategies</li><li>The implications of poor code</li><li>The importance of back &amp; forward testing</li><li>Strategy Optimisation</li></ul> I also realized that whilst coding algorithmic algorithms is both fun and almost quite addictive, I found that in most cases the only strategies that can really be bulletproof against absolutely all market conditions are ones that are able to constantly adapt to the current market conditions. these more flexible EAs usually involve a range of indicators running off of various different confluences and market cycles and will use the appropriate trading strategy depending on the market conditions.Likewise, many strategies also lend themselves to certain currency pairs or timeframes. I found that my best strategies and indicators had to take into account all time periods regardless of if we were only placing trades in one of them. This was an important factor in fine-tuning entry and exit strategies to employ sensible risk margins and leverage. I also found that aside from complex algorithmic strategies, in general strategies using machine learning such as neural networks tended to outperform all other strategies in the long run. In my opinion, this is due to their ability to adapt to market conditions effectively much like what I was trying to achieve with multi-strategy EAs. My most successful EA was my 'mothership' EA which could be configured to work with a range of market conditions.<ul><li>Confluences could be added to form a single strategy such as:</li><li>Multi timeframe support &amp; resistance</li><li>multi timeframe trend lines</li><li>MA's, MA Cross, &amp; MA ribbon</li><li>Aroon &amp; alligator for longer market changes</li><li>Bollinger bands for unstable markets</li><li>Heiken Ashi candlesticks to reduce fakeouts &amp; work off of more sensible averages in price data</li><li>parabolic SAR for timed entries exits on trend following strategies</li><li>momentum indicators for lower timeframe changes to spot reversals</li><li>candlestick pattern recognition for entries &amp; exits as well as reversals</li><li>RSI &amp; Stochastic indicators,</li><li>Divergence indicators that can be applied to other indicators against the price action trend lines</li></ul> Through creating these various strategies and functions I found that the divergence indicator was very effective when combined with other confluences, especially if it was run on all timeframes concurrently to spot exact points in time when all timeframe signals were aligned. As this was added on larger timeframes such as the hourly, 4hourly &amp; daily it would be a rare occurrence to get signals aligned across all timeframes but it did hugely increase the edge in the market when it happened. Working with lower time-frames also allowed for much more fine-tuned entries which allowed for very tight stop losses resulting in a strong risk-reward ratio. I added a lot of extra features to my EAs such as:<ul><li>Hedged trading</li><li>Stop and limit orders,</li><li>auto lot-sizing</li><li>multiple trades</li><li>stop losses and trailing stops</li><li>The option to add extra strategies on top such as martingale (although I don't recommend doing this).</li></ul>

I created an email subscribe form as an npm package to integrate with AWeber (an email newsletter management service). This package has been used on various sites such as howtobet.com

This is a NodeJS project with no frontend that I designed to run on a server to slowly increase a user's followers by following and liking posts, profiles, and hashtags at a set schedule. All parts of the bot are configurable to determine how and when it goes about its daily tasks. I used Puppeteer along with some npm packages and evasive techniques to make it harder for Instagram to detect the bot.The bot was fairly successful however using more aggressive settings occasionally caused the account to get flagged, I have since been made aware of newer packages which have been developed specifically to evade detection and mock different user agents so as to spoof a typical user on their device. I added extra functionality to allow the user to input certain handles so that the bot would predominantly target those handles, hashtags, and similar accounts. with extra options for liking, following, and setting like/follow ratios as well as scraping handles from accounts.

I was part of a 2 man dev team that developed howtobet.com from the ground up. HowToBet is a React &amp; NextJS application using WordPress as a backend CMS. It is quite a feature-rich platform, utilizing geolocation IP tracking to show different information and promotional codes to users depending on what US state they are visiting to comply with US State law. This site uses a custom node express server in the frontend NextJS framework as well as more complex components such as state variable toplists and live odds comparison feeds. To make the development experience easier for other devs to work on we dockerized both the backend and frontend repositories of this project.

Bad Syntax is a personal project where I created a website to propose a stage concept to Boomtown Festival in 2020. The idea of the stage was centered around artificial intelligence and overconsumption of data, so the site leverages data collection, which would then be used as part of the stage visuals. This project was centered around a hybrid of creativity mixed with technology, so aside from some pretty WebGL shader effects, I incorporated some fun functionality into this system such as the face-morph video, a type of StyleGAN, (Generative Adversarial Network) which is video encoded on the site for demo purposes but was intended to be encoded to DXV3 for life for stage visuals. I used face-API and a few other libraries to create the web camera face detection system, which essentially stores facial features as a fingerprint to later recognize the same face. The system can also display dynamic face detection boxes around a person along with a percentage of certainty that it has matched the data points. Further information can also be displayed such as an approximation of the age and gender of the subject. I also added some web crawlers to the site with a social search API. these are used for code display &amp; data collection. A user enters their name into the input and the site crawls websites such as Google and popular social media platforms to try and find information on the user to display in the form of a JSON document that automatically scrolls the page. The main site itself is a node server running express. File upload &amp; storage is to Amazon AWS and EJS has been used as the frontend templating engine. Any user data is saved to MongoDB. this was done so the data could easily be retrieved in real-time for display on the screens. My plan for the image manipulation was to use an AWS pipeline consisting of the input images, some Lambda functions running the face morph AI and then finally FFmpeg for video encoding and either streaming the returned video to a destination or saving it to an S3 bucket.

This site is a subscription-based crypto signal site I did for a client. The site is built in Express, using EJS as the frontend templating engine for view rendering. It utilized the CoinApi API for fetching cryptocurrency data and manipulating that data into signals prior to saving it to a MongoDB cluster. The platform was pay-walled using Stripe JS, set up for taking subscriptions. WebSockets were used for the signal data and for future plans to have an open WebSocket connection displaying the generated signals and OHLC data on a candlestick Chart. This site is no longer in production but I have kept a snapshot of the repository.

This platform is a business venture co-founded by me and my brother. This originally started as a Node and Express platform with AWS, however, we needed a more user-friendly CMS for my brother and creators to use as the project grew so we switched it to WordPress. This site is quite media heavy so we had to spend some time strategizing how to best serve content such as image thumbnails, gifs, and video content.

I trained and managed a 5 person Ukrainian development team.I went to Odessa in Ukraine on two separate occasions to conduct a week of intensive training on WordPress &amp; NodeJS, I then managed this team for approximately 6 months. As a team, we overhauled approximately 150 WordPress affiliate sites that were very outdated or were acquisitions from other companies that needed to be brought in line with our company standards. Approximately a year after the first management task, I later re-trained this team on React, Redux &amp; NextJS with a headless WordPress backend. I conducted this training using Microsoft Teams and recording instructional videos during the covid pandemic.

USB Sessions is a simple WordPress blog site I did for a friend with a video stream on the homepage for when he is live streaming DJ mixes.

I was tasked to create a custom WordPress theme for easily launching new iGaming casino sites in WordPress. I built the theme using Bootstrap and Sass, creating a set of useful casino components and post-type layout templates that could be used in a modular way. This core theme was used across approximately 20 sites and maintained centrally by me in a single repo using GitLab runner.

I was the lead developer on Cryptimi.com, a Cryptocurrency affiliate and blog website owned by an iGaming company. This WordPress site utilises a custom theme.

Slave Studios is a personal business project I created using WordPress.

I was part of a development team administrating around 150-200 websites. The sites listed are the ones I had the most active role developing. Tasks on these sites ranged anywhere from frontend UI to building custom themes &amp; plugins, full site overhauls, and building REST APIs. A lot of the work on these sites was very SEO orientated due to the nature of the iGaming industry.

Music Industry Hosting is a personal business project. It is a platform I set up to provide reseller hosting using WHM and WHMCS. this system allows me to sell, administrate and manage web hosting along with domains and cPanel accounts. This was predominantly used for friends &amp; family as well as smaller WordPress websites.

Studioslave is a personal business of mine. It is a learning &amp; eCommerce platform for electronic music production, built with WordPress and hosted as part of a traditional LAMP stack using cPanel with S3 integration for storage. It made use of the Vimeo API to allow for streaming video content as part of paid courses. The platform was initially launched on EC2 but I have since migrated it to a cheaper solution with less management requirements as it is no longer actively maintained.

Having used AWS extensively as a developer over the past 5 years. I recently spent some of my personal time to up-skill and get certified as an AWS Solutions Architect Associate in April 2023. As a certified AWS Solutions Architect Associate, I am competent in: <ol><li>Architecting and designing secure, reliable, and scalable AWS solutions based on clients' requirements, utilizing AWS services and best practices.</li><li>Estimating AWS costs and identifying cost-control mechanisms to help clients minimize expenses while maintaining performance and reliability.</li><li>Recommending and implementing appropriate data storage solutions, such as Amazon S3, Amazon EBS, Amazon RDS, or Amazon DynamoDB, based on clients' requirements and access patterns.</li><li>Selecting and configuring appropriate compute services, such as Amazon EC2, AWS Lambda, or Amazon ECS, to meet the performance and scalability needs of applications.</li><li>Designing and implementing secure and scalable networking solutions using AWS services like Amazon VPC, AWS Direct Connect, Amazon Route 53, and AWS Global Accelerator.</li><li>Ensuring data security and compliance by implementing AWS security services and features, including AWS Identity and Access Management (IAM), Amazon Cognito, AWS Key Management Service (KMS), AWS WAF, and AWS Shield.</li><li>Assisting clients in migrating existing on-premises or cloud-based applications to AWS and integrating AWS services with existing systems.</li><li>Designing and implementing monitoring, logging, and auditing solutions using AWS services like Amazon CloudWatch, AWS CloudTrail, and AWS Config to maintain the health and performance of the infrastructure.</li><li>Developing and implementing high availability and disaster recovery strategies using AWS services like Amazon RDS Multi-AZ, Amazon S3 Cross-Region Replication, Amazon Route 53 Health Checks, and AWS Backup.</li><li>Continuously staying up-to-date with the latest AWS services, features, and best practices to improve existing solutions and provide better recommendations to clients.</li></ol> By possessing these competencies, I can help clients build efficient, secure, and scalable solutions on the AWS platform.

AWS Solutions Architect

I have built chrome extensions that utilise both React and AWS Serverless architecture written in typescript.

Chrome Extension Development

I have harnessed the power of OpenAIs API's such as Whisper, Dall-E and ChatGPT Chat &amp; Completions to build serverless applications, bots and chrome extensions. I have stacked these API's together in some projects for more obscure use cases and to modify the likes of ChatGPT to more niche use cases. I am also signed up as a beta member for ChatGPT to experiment with new features.

ChatGPT & OpenAI APIs

Pinia is a new and improved state management system for Vue and Nuxt applications. Many have dubbed it as as the successor to Vuex. I used Pinia when converting an existing project from Nuxt 2 and Vuex to Nuxt 3 and Pinia. I found Pinia to be very simple to setup and use and much the same as Vuex and Redux.

Pinia

As a user of the React Hook Form npm package, I excel in implementing efficient form validation and state management in React applications, integrating with popular UI libraries, and handling complex form components. I am skilled in managing dynamic form fields and utilizing React Hook Form DevTools for debugging during development.

React Hook Form

I learned Vuex along with Vue and Nuxt for a project with Raketech. Whilst i find the Vue, Vuex &amp; Nuxt system extremely easy to use, i prefer using react with redux purely because react is my frontend framework of choice.

Vuex

Vuetify is a front-end material design CSS framework. I started using Vuetify when building a content management system as part of a consultation project. It is a great framework to use for building front-end interfaces due to its ease of use and well-designed API which is in the usual material UI format and therefore very easy to implement across multiple languages such as Vue (Vuetify) and React (Material UI/Minimal UI).

Vuetify

I added VueJS to my skillset as it was a requirement for a consulting project that was using Vue, Vuex, Vuetify Nuxt, and Firebase. Whilst I like the ease of use and speed of development with Vue's class/object-based approach. I prefer the freedom that React offers with the ability to have more fine-grained control over the rendering cycle. learned Vue via the Vue documentation, online tutorials, and a Udemy course by Maximillian Schwartzmueller. I recently upgraded a site fro Vue 2 to Vue 3, so have experience in both versions as well as the upgrade process.

Tailwind is a front-end CSS framework heavily centered around the design principle of using attribute classes in HTML markup rather than component-specific classes when creating components. I think it is a great tool to use when building component libraries with reusability in mind. I have used tailwind CSS and Typescript for a simple nextJS website for a pub.

Tailwind Css

Algolia is a cloud-hosted search &amp; discovery platform enhanced with AI. It integrates with your backend database and is kept in sync using CRUD operations. it is extremely efficient for tasks such as indexing, complex queries, search, and recommendations. It is great technology when dealing with SEO-reliant or e-commerce-style websites that need fast &amp; flexible &amp; indexing, queries, and recommendations of products or records within a catalog, I have used Algolia on HowToBet.com as well as using it extensively on Audiolancer.io to ensure MongoDB models were synchronized with Algolia via MongoDB middleware functions. I used middleware to ensure that the integration was as seamless as possible whilst requiring very little extra work during continual development due to its close coupling to database CRUD operations. The Audiolancer usage of Algolia also involves some fairly complex facet filtering to ensure records can be easily queried by any attribute, intelligently highlighted, sorted, and displayed to my specification.

Algolia

I have used both TDD &amp; BDD testing when using tools such as Cypress &amp; Jest as well as writing tests in other systems such as Storybook &amp; Postman. I have written a full range of tests from E2E, integration &amp; unit tests for maximal code coverage. Prior to using Cypress &amp; Jest I has some limited experience using Mocha, Chai &amp; Sinon which are essentially the building blocks of more fully fledged testing suites such as Cypress.

BDD & TDD Testing

Cypress is a browser-based javascript framework for testing web applications. It can be used for both BDD &amp; TDD testing styles. It is my testing library of choice and I have used Cypress extensively for integration and e2e testing of audiolancer.io as well as API integration tests for a consulting client. Through using cypress in my projects, I have written hundreds of tests, covering APIs, simple functions, full e2e flows, and payment systems.I am competent at writing tests that are:<ul><li>effective</li><li>don't break on superficial code changes</li><li>not prone to flakiness such as intermittent and difficult-to-debug test failures</li></ul> Some more advanced features of Cypress that I incorporate into my tests are: Hooks, Interceptions, mocks, stubs, spies, tasks, fixture, commands, seeding test environments &amp; teardown of test environment for each test/spec, testing authentication flows with sessions and cookies, Testing email send-outs and parsing responses, Testing social logins with Puppeteer and lastly, Refactoring common testing functions into commands, tasks, and plugins. I have also got experience with adding Cypress to a CI/CD pipeline in Gitlab.This involves ensuring the CI environment is set up correctly (and preferably split into stages if necessary) to ensure all tasks within a pipeline can be completed. Sometimes environments need to be customized to deal with extra packages involved in testing such as puppeteer.Adding Cypress in a CI pipeline like this has been fundamental to ensure code is carefully scrutinized prior to ending up on a staging or production environment and that deployments fail in the case of any failing tests. I have also integrated Jest into a CI pipeline along with code instrumentation so that I can get merged code coverage reports in various formats produced as an artifact at the testing phase. These reports comprise of a full frontend and backend coverage report of my Audiolancer platform showing all tests (e2e, integration &amp; unit tests) and any failing tests will have screenshots and videos saved as an artifact for later review.

Cypress

I have been using material UI as my main frontend CSS framework for personal React projects since moving away from react-bootstrap and Reactstrap. The reason for this is to reduce development time whilst still being able to produce an extremely well-polished and professional-looking frontend. Furthermore, the material UI framework API is fairly generic across multiple languages, making it easy to build my own personal component library enabling me much faster integration of frontend interfaces with tooling I commonly use such as state management systems like Redux &amp; Vuex.

Material UI

Nuxt JS is a front-end Vue framework. It is the Vue equivalent of NextJS for React and It is a great tool for building websites with SEO in mind. It has the same capabilities as NextJS in terms of building powerful applications that scale and can be rendered either on the client or server.I learned Nuxt for a project whilst consulting for one of my main clients and I found it almost identical to NextJS. I prefer to use NextJS and React purely because I have more experience with NextJS due to using it daily since version 8. I also feel that NextJS is slightly more mature with how it splits API routes into lambda functions when deployed to Vercel. (we have had issues with Nuxt projects hitting lambda limits due to having too many npm packages in a single project). whilst this issue is likely to be corrected in the future, for this reason, I prefer NextJS. I have recently worked with the latest version of Nuxt (version 3 as of writing this). Which I have found to be a huge improvement from Nuxt 2 despite a huge paradigm shift in many of Nuxt's features, with the most notable being the change to the Nitro &amp; H3 server, event handler syntax, and middleware functionality.

NuxtJS

I started learning Typescript as part of my personal development goals by incorporating it into my smaller personal projects such as my ThreeJS vinyl spinner project. In mid-2021 I started the process of overhauling my larger Audiolancer.io project to benefit from Typescript. Due to the sheer size of the Audiolancer platform, this upgrade is likely to take me well into 2023. I have some previous experience with using strongly typed languages such as MQL4 and MQL5 (MetaQuotes Query Language) which is code for writing Forex and Crypto bots. Much like Javascript, it is written on top of C++, however, MQL4 and MQL5 are slightly lower-level than Javascript and offer a lot fewer features.

Typescript

ReactStrap is a frontend library or prebuilt Bootstrap 4 components written in React. The library is built with speed, flexibility, and ease-of-use in mind. I started using Reactstrap in my personal Audiolancer project as a replacement for react-bootstrap. I have since moved towards material UI as my go-to frontend styling library but still actively maintain projects that use Reactstrap.

Reactstrap

ThreeJS is a 3D rendering library utilizing WebGL and written in Javascript. I learned ThreeJS for a personal project which uses it along with React &amp; Typescript. Whilst I'm fairly competent with ThreeJS, it is quite a vast eco-system that requires a strong skillset in areas such as coding, design, and visuals,I am not an expert in ThreeJS as it is not a part of my daily coding arsenal, however, I am always up for furthering my skills in this field with new projects.

ThreeJS

Storybook is&nbsp;a front-end workshop for building UI components and pages in isolation. I have used it on one of my consulting projects where we used it to build a fully tested and documented component library with React &amp; Typescript. I found it to be great in terms of having a clean and friendly GUI for documenting and testing frontend component states. I intend to use Storybook more in the future when time allows.

Storybook

MetaQuotes Language 5 (MQL5) is&nbsp;the integrated programming language for developing indicators and trading strategies for the free-of-charge online trading platform named MetaTrader 5. Over a hundred brokerage companies and banks use the platform to provide their services to customers. I taught myself how to write EAs (trading bots) using MetaTrader 4 &amp; 5by creating a variety of automatic forex trading algorithms over the course of 8 months, ranging from Support and resistance strategies, Trend lines, RSI &amp; stochastics, Divergences, Ma ribbons and combinations of all of these indicators and systems combined. I deployed these EAs to servers, back-tested and forward-tested the strategies, optimized the input parameters, and logged results in MyFXBook.I learned a lot about time-series data, writing MQL4 /MQL5 code, and the benefits &amp; limitations of writing code using strongly typed languages derived from C++.

MQL4 & MQL5

Stripe is&nbsp;a cloud-based solution that offers a varied set of unified APIs and tools that enable businesses to manage and accept payments online. Stripe handles key payment functions, such as subscriptions, storing cards, and direct payouts to user bank accounts. I use StripeJS extensively to take payments on my Audiolancer.io platform as well as my Crypto signal app. There are multiple payment systems on Audiolancer which cater to a range of site requirements such as: <ul><li>Acting as 'escrow' for held funds until a service has been rendered to a satisfactory standard</li><li>Instant withdrawals</li><li>Payments between users using 'stripe connect'</li><li>Multiple &amp; split vendor payouts</li><li>Stripe transfers</li><li>Extensive use of Stripe webhooks</li><li>Stripe subscriptions using pricing and products</li></ul> I think stripe is an extremely powerful, innovative, and developer-friendly system for payment processing and one that I see myself continuing to use for the foreseeable future.

Stripe Payments

Docker is&nbsp;a software platform that allows you to build, test, and deploy applications quickly. Docker packages software into standardized units called containers that have everything the software needs to run including libraries, system tools, code, and runtime. I have used docker for development environments whilst working as a team on sites such as howtobet.com and casinoguide.se. I have also done a course on how to work with docker and Kubernetes to create custom images for local development tasks. I have also found docker containerization to be a useful skill to know when it comes to running tests in a CI environment such as GitLab as well as for other tasks such as creating and running Lambda functions in AWS that require lots of dependencies where the use of Lambda layers may not be practical.

Docker

Hooks are&nbsp;functions that let you “hook into” React state and lifecycle features from functional components. I have been using hooks since they were first publicly introduced. Asides from useState and useEffect, I have also spent some considerable time learning hooks such as:<ul><li>useRef</li><li>useMemo</li><li>useCallback</li><li>useLayoutEffect</li><li>useContext</li><li>My own custom hooks</li></ul> I write custom, reusable React hooks for projects that have multiple components with often repeated code that could easily be refactored.Common examples of useful hooks are within form components, useDarkMode, useFetch, useIsMounted, &amp; getting and setting web APIs such as localStorage. In some cases I use the useContext hook for simple state management tasks such as general settings &amp; dark mode, however, most projects I work on are quite complex and therefore easily managed with state management systems such as Redux

React Hooks

TensorFlow.js is&nbsp;a JavaScript library for training and deploying machine learning models in the web browser and in Node.js. I have learned and used TensorFlowJS via some online courses and doing a personal portfolio project to demonstrate machine learning skills with image classification &amp; processing by building an automated bot around the popular 'Tinder' dating app web API. The app exposes a set of functions for:<ul><li>analyzing profiles</li><li>scraping bios</li><li>automatic intelligent swiping based on your previous swipe history</li><li>geolocating to new locations</li><li>replying to messages</li></ul> This code was developed as an interesting project to demonstrate the uses and dangers of machine learning and has not been made publicly available. I have not written a huge amount of TensorFlow or machine learning code outside of this project, however, I do have an active interest in how the technology develops and I am always willing to accept new projects utilizing this technology so that I can further develop my knowledge in this area.

TensorFlowJS

Puppeteer is&nbsp;a Node library that provides a high-level API to control&nbsp;headless&nbsp;Chrome or&nbsp;Chromium&nbsp;over the DevTools Protocol. It can also be configured to use full (non-headless) Chrome or Chromium, enabling you to control Chrome from the command line. Puppeteer can be used for&nbsp;web scraping, automation, and testing purposes. I have used puppeteer a lot for browser-based automation tasks. specifically for creating bots for social networking such as Instagram and Tinder.

Puppeteer

Jest is a JavaScript testing framework designed&nbsp;to ensure the correctness of any JavaScript codebase. It allows you to write tests with an approachable, familiar, and feature-rich API that gives you results quickly. Prior to Jest I had some experience with Mocha, Chai, and Sinon and I have since moved over to using Jest &amp; React Testing Library for unit tests, and Cypress for e2e, API &amp; integration tests. Some slightly more complex tasks I have done with Jest involve writing higher-order components for wrapping and mocking redux providers &amp; stores during unit and integration testing. For React I use React testing library.

Unit Testing With Jest & RTL

Through various personal projects, I have worked with many of the browser's web APIs and objects such as: canvas, web push, web audio, service workers, clipboard, cookies, local storage, session storage, geolocation, beacon, navigator, WebSockets, WebGL, streams &amp; media recorder to name a few.

Web APIs

Redux is&nbsp;a predictable state container designed to help you write JavaScript apps that behave consistently across client, server, and native environments, and are easy to test. While it's mostly used as a state management tool with React, it can be used with any other JavaScript framework or library. Redux is my go-to state management system. I have used different variations of it (both redux and redux toolkit) as well as extra modules such as redux-thunk for asynchronous functions, and redux dev-tools. I have used redux on multiple headless WordPress projects as well as most of my NextJS projects. I have also spent time training a team of Ukrainian developers on how to use Redux within a React and NextJS stack.

Redux

Docusaurus&nbsp;allows developers to use tools they already know like Markdown or MDX to write documentation or blogsIt can be used as an npm package that builds a directory and displays a simple documentation page that runs as a server. I have used it in a project to keep better documentation on how to administrate and develop Headless NextJS applications, as well as when maintaining documentation and useful training course information for new developers joining the company I do consultation work for.

Docusaurus

JSDoc is a commenting strategy and documentation generator which creates docs from reading formatted comments above function definitions. I used JSDoc on HowToBet.com and a few other projects as my commenting strategy to make functions in a shared repo easier for other devs to use. It helped me realize why languages such as Typescript are so useful due to their added advantage of auto-documentation of params, types, and return values without having to write any extra code or documentation. In general, I still like to write strong descriptions and clear documentation where time allows for ease of continued development, especially when operating as part of a larger team.

JSDoc

Postman is a tool for developing, documenting, and monitoring REST APIs. I use it for both personal and work projects to ensure my endpoints are written to OpenAPI (3.1) specification, are well documented, and are easily shareable with other team members. I also find Postman is a great tool for visualizing API flows, writing pre-request scripts (which aids in testing authentication), and also writing simple API route tests within the postman interface, however, I prefer keeping tests to Cypress where possible.

Postman

Git is&nbsp;a DevOps tool used for source code management. I have been using Git for all personal and business projects with a focus on maintaining a good git-flow when working in a team. In general, I work on a feature branch, then merge that branch with the dev branch, which in turn can be merged with staging and master.both staging and master branch merges will automatically trigger a deployment. For deployments and all things CI/CD related I predominantly use Gitlab, however prior to GitLab, I used GitHub for a few years. GitLab is my main version control provider of choice because of how well it works with Vercel deployments and its ease of use when it comes to configuring pipelines, jobs &amp; schedules.

Git Version Control

As a developer with a lot of personal and business projects, I have always had an active role in managing my own deployments and CI/CD. Prior to moving away from the LAMP stack, this was done via pushing code to GitHub and GitLab repos, but since moving to full-stack Javascript, I now use GitLab and Vercel for deployments, automation, testing, and any other CI/CD tasks. In terms of DevOps pipelines. I have a firm understanding of how systems should be built to be both flexible and scalable as well as ensuring they are easily manageable. When I am in control of setting up a system I will always try to keep the pipeline process simple and easy to understand whilst still having clear segregation between environments. For me, a typical system should consist of a single repo with multiple branches for Git Flow. Each branch should be connected to a deployment accordingly (dev, staging, production). and each environment should have a separate database within the same DB cluster, with an extra DB instance for testing which can be easily switched to using node environment variables at runtime or via ci input. I also have experience using Docker to create development and testing environments. On my personal projects, I find there is no requirement to use Docker as I am solely developing projects on my own and it will not be run within docker once deployed. I also do a wide range of other DevOps tasks, such as database migrations, syncing of AWS assets (s3 content) between buckets and between local machines, and Algolia search management using various command line tools such as the Vercel CLI, AWS CLI, MongoCLI and Algolia CLI.I also have experience with working with AWS SAM and CloudFormation to save my infrastructure as code.

DevOps & CI/CD

SASS (Syntactically Awesome Stylesheets) is a CSS pre-processor that&nbsp;lets you use variables, mathematical operations, mixins, loops, functions, imports, and other interesting functionalities that make writing CSS much more powerful. I learned Sass whilst fully employed with the company I now consult for as a freelancer. I've used it extensively throughout most of my projects, It is my CSS styling system of choice when not using external libraries or when augmenting CSS framework libraries with custom styles. The reason I prefer it over standard CSS is that I find the code is cleaner and has many advantages over regular CSS such as with variables and scoped nesting.

Sass

Eslint &amp; Prettier are useful coding tools to enforce more opinionated coding styles and formatting. I have been using these since moving to the MERN and MEVN stacks. Whilst these aren't really skills, i think it is good to have working knowledge of how these tools work in a team environment and how they can benefit a project by ensuring everyone is adhering to the same coding standards and can also get coding hints as to why certainissues may arise and how to avoid them in the future.

Eslint & Prettier

OpenAPI is a specification for writing REST APIs which can be easily understood by humans whilst also being machine readable. I have recently focused time on ensuring APIs I am actively developing adhere to OpenAP 3.1 standards to ensure they can be more easily consumed. by developers and end users alike.

OpenAPI

I learned socket.io for a crypto/forex related JAM stack application. The aim of using socket.io was for sending and receiving crypto signals as well as getting timeseries OHLC candlestick data with a single connection which stays open instead of a rest API being repeatedly polled.

Socket IO

The WebSocket API is an advanced technology that makes it possible to&nbsp;open a two-way interactive communication session between the user's browser and a server. With this API, you can send messages to a server and receive event-driven responses without having to poll the server for a reply. I used plain WebSockets to embellish my Audiolancer.io platform messaging system. Originally sent peer-to-peer messages were saved to the database and would require the receiving user to refresh their page to have the new message fetched from the database. Whilst keeping this system in place, I added a WebSockets layer that implements AWS DynamoDB and some serverless AWS Lambda functions to allow for an instant messaging experience between users of the platform with no requirement to refresh the page, similar to applications such as Slack.

Web Sockets

Next.js is a flexible&nbsp;React framework&nbsp;that gives you building blocks to create fast&nbsp;web applications. NextJS makes the building of React Apps faster and more efficient by streamlining development tasks, such as: <ul><li>User Interface&nbsp;- how users will consume and interact with your application.</li><li>Routing&nbsp;- how users navigate between different parts of your application.</li><li>Data Fetching&nbsp;- where your data lives and how to get it.</li><li>Rendering&nbsp;- when and where you render static or dynamic content.</li><li>Integrations&nbsp;- what third-party services you use (CMS, auth, payments, etc) and how you connect to them.</li><li>Infrastructure&nbsp;- where you deploy, store, and run your application code (Serverless, CDN, Edge, etc).</li><li>Performance&nbsp;- how to optimize your application for end-users.</li><li>Scalability&nbsp;- how your application adapts as your team, data, and traffic grow.</li><li>Developer Experience&nbsp;- your team’s experience building and maintaining your application.</li></ul> I personally use NextJS for building all kinds of web applications. I am a huge advocate of NextJS and have used it since version 8. I have used it extensively for static websites, hybrid sites, e-commerce platforms, and sites with specific requirements when it comes to SEO such as choices in client side or server rendering. I have a deep understanding of how NextJS works including features in newer versions such as middleware and edge functions. I have trained a team of Ukrainian developers on using NextJS, React, and Redux. Next is a core part of my day-to-day development work as I feel it is at the forefront of the future of the web. Notable projects that I have used NextJS with are Casinoguide.se, Howtobet.com, and Audiolancer.io

NextJS

FFmpeg is an open source library for recording, streaming, converting, and editing video, images, and audio. It comes in many formats such as a command line tool, web server, or wrapped as a javascript library. I have used it in NodeJS applications for backend video conversions for a DJ Visual Loops website https://djvisualsbuilder.com/ as well as in a personal project for converting videos created from a node canvas in the browser on the fly. I have also previously used it as an AWS Lambda function for transcoding audio between .wav format and .mp3 format to reduce the size of files streamed to the front end of my audiolancer.io music services website, although recently I replaced this FFmpeg Lambda for Amazon AWS's elastic transcoder service which is essentially the pre-packaged and easily configurable AWS version of the FFmpeg transcoder.

Ffmpeg

React is a&nbsp;JavaScript&nbsp;library created by&nbsp;FacebookReact is a&nbsp;User Interface&nbsp;(UI) libraryReact is a tool for building&nbsp;UI components React is also my favorite library for writing frontend components. I Initially learned React using legacy class-based components, but I have since up-skilled to use the more modern functional components.I have spent a lot of time learning the more complex parts of React such as:<ul><li>performance optimization with the react profiler to reduce slow or unnecessary re-renders</li><li>using more complex and lesser-understood built-in react hooks as well as custom ones</li></ul>

React

I have used object oriented programming across both PHP &amp; Javascript. In javascript it works slightly differently with the use of prototypal inheritence. I used OOP heavily on my "tinder api" project, but in general, for frontend tasks, I write functional code to stay in line with Reacts move to functional components. For backend tasks I am able to write in an OOP style if required but also tend to lean to functional code within API routes that operate directly on a database so there is little need to write using OOP when properly adopting the MVC model.

Amazon Simple Storage Service (Amazon S3) is an object storage service that offers industry-leading scalability, data availability, security, and performance. You can use Amazon S3 to&nbsp;store and retrieve any amount of data at any time, from anywhere. I have used S3 for public static asset folder hosting for things such as image assets, video assets, and sitemaps updated via cron jobs as well as for user-generated content. I have also used S3 for hosting static websites, public content, and more niche uses such as streaming audio with complex bucket policies a well as user privilege-locked content with a tiered storage setup.

AWS S3

Amazon Elastic Transcoder is&nbsp;media transcoding in the cloud. It is designed to be a highly scalable, easy-to-use, and cost-effective way for developers and businesses to convert (or “transcode”) media files from their source format into versions that will playback on devices like smartphones, tablets, and PCs. I use elastic transcoder on Audiolancer.io for transcoding files uploaded to S3 buckets between .wav and .mp3 to ensure I have multiple formats available for different tasks such as downloading vs streaming to save bandwidth, increase performance and reduce data transfer costs.

AWS Elastic Transcoder

Amazon Simple Notification Service (Amazon SNS) is a managed service that&nbsp;provides message delivery from publishers to subscribers&nbsp;(also known as producers and consumers). Publishers communicate asynchronously with subscribers by sending messages to a topic, which is a logical access point and communication channel. I use AWS SNS Predominantly for topic subscription and SMS messaging (bulk send-outs &amp; singular automatic verification SMS messages) for Audiolancer.io

AWS SNS

I have been using and learning AWS's arsenal of products since 2018, and have undertaken the Certified Developer Certification and Architect Associate Certification courses on Udemy as well as had a host of hands-on experience integrating AWS services into my personal &amp; business projects. Whilst I find the AWS Javascript SDK V3 documentation to be sub-standard, (as it is a fairly new version and quite poorly documented). I do find myself using AWS services very regularly due to the robustness that these enterprise-level products provide. Whilst I am always learning new things and developing my skillset within the AWS ecosystem, I have a firm understanding of the majority of the most common services from both an architectural perspective as well as the perspectives of administrating the AWS account, and implementing the service on an application as a developer. AWS services I often use are: <ul><li>AWS S3 for storage of user assets</li><li>Cloudfront for keeping resources available across multiple availability zones at lower latency</li><li>Budget Manager as a failsafe for ensuring there are no critical issues with the account that would affect the budget</li><li>elastic beanstalk for provisioning assets such as EC2 servers</li><li>Cloudwatch &amp; CloudTrail for monitoring other services and keeping logs (specifically for Lambda functions)</li><li>EC2 for running servers however in recent years I have moved away from having dedicated servers to the much more cost-efficient use of serverless functions</li><li>SNS and SQS for queuing messages and dead letter queues as well as topic subscriptions and SMS messaging</li><li>SES for email send-outs, templating, and customer list management.</li><li>Route53 for managing domains.</li><li>Cognito for managing AWS users and issuing access credentials scoped to specific S3 prefixes and methods via IAM roles.</li><li>I have since moved away from Cognito in favor of more manageable systems for NextJS applications such as nextAuth (due to easier database access).</li><li>IAM for managing AWS roles and permissions.</li><li>AWS Lambda for creating serverless functions triggered by both the javascript SDK and API Gateway.</li><li>Eventbridge for running cron jobs and scheduled events on websites without affecting the website's NodeJS event loop via lambda functions</li><li>Elastic transcoder for creating automatic audio transcoding pipelines within an S3 bucket from .wav to mp3.</li><li>AWS SAM for local development of Lambda functions and lambda layers as well as for templating the Lambda infrastructure as code.</li><li>I also use AWS cloud formation for creating and keeping logs of my AWS infrastructure</li></ul>

AWS Amazon Web Services

MongoDB is&nbsp;an open-source NoSQL database management program. NoSQL is used as an alternative to traditional relational databases. NoSQL databases are quite useful for working with large sets of distributed data. MongoDB is a tool that can manage document-oriented information, and store or retrieve information.MongoDB is a powerful NoSQL (schemaless) document database that is highly scalable. It is my database of choice and I use it extensively in all suitable projects. I use a lot of the more advanced MongoDB features such as: Enforced schemas, Document embedding &amp; population, Custom validation, Pre and post-middleware, Static model functions, Aggregation pipelines &amp; Virtual properties. I usually use MongoDB as a cloud instance in MongoDB Atlas and connect &amp; administrate the instance with tools such as the MongoCLI command line tool or MongoDB Compass. Recently I have started using the AWS equivalent to Atlas, which is called 'DocumentDB'.

MongoDB

WHM provides a set of features to manage your cPanel server, its services, and its accounts.&nbsp;WHMCS provides automated functionality for account management, including billing services.&nbsp; I used these tools to automatically provision clients with hosting and domains. I did this from my website www.musicindustryhosting.com

Whm & Whmcs

Amazon SES is a cloud email service provider that can integrate into any application for&nbsp;bulk email sending. Whether you send transactional or marketing emails, you pay only for what you use. Amazon SES also supports a variety of deployments including dedicated, shared, or owned IP addresses. I AWS Simple Email Service for sending verification emails, bulk email send-outs, list management, and holding email templates for Audiolancer.io and email campaigns for StudioSlave.com

AWS SES

Ecmascript 6. Learned as a core part of various boot camps and courses taken on Udemy.Used across all of my Javascript projects since learning it.

API Gateway&nbsp;acts as a "front door" for applications to access data, business logic, or functionality from backend services, such as workloads running on Amazon Elastic Compute Cloud (Amazon EC2), code running on AWS Lambda, any web application, or real-time communication applications. I have used AWS API Gateway for building REST APIs &amp; websockets messaging apps tied to lambda functions. Whilst i think it is a great service for writing SAAS cloud-based APIs and microservices, I do not need to use it that often because frameworks such as NextJS, NuxtJS and other platforms that use Vercel are already leveraging AWS serverless functions to run API endpoints in as serverless functions on AWS and at the edge.

AWS API Gateway

Mongoose is an Object Data Modeling (ODM) library for MongoDB and Node. js. It&nbsp;manages relationships between data, provides schema validation, and is used to translate between objects in code and the representation of those objects in MongoDB Essentially, MongooseJS is a javascript driver for MongoDB and it exposes a more user-friendly API for writing code to interact with the database within a javascript node server environment. Whilst I do still write standard MongoDB queries from time to time (or for more complicated queries), in general, I prefer to use Mongoose because of its simplicity and intuitive API. I think it is a great tool for adding object modeling schema and validation to database models.

Mongoose

AWS Lambda is a serverless compute service that&nbsp;runs your code in response to events and automatically manages the underlying compute resources for you. These events may include changes in state or an update, such as a user placing an item in a shopping cart on an eCommerce website. I love Lambda due to its cost-efficiency and scalability with little extra work. I use it extensively in multiple projects such as My Online Pdf Contract Create &amp; Sign App, Audiolancer.io, and Picks &amp; Parlays. I have used Lambdas for a wide range of tasks such as for running tasks on S3 buckets (using s3 as the trigger), running hourly, daily, and monthly cron jobs (using EventBridge), executing via API Gateway, and also executing directly from code, both synchronously and asynchronously for tasks such as API route handlers, file compression, background tasks and chat systems using WebSockets. I also use lambda layers to allow myself to reuse dependencies and modules across my lambda functions.

AWS Lambda

Amazon CloudWatch is&nbsp;a monitoring and management service that provides data and actionable insights for AWS, hybrid, and on-premises applications and infrastructure resources. I use CloudWatch along with CloudTrail to monitor all of my other AWS Services and to generate logs and reports to give greater insight into the usage and statistics of my AWS account.

AWS CloudWatch

Amazon CloudFront is&nbsp;a web service that speeds up the distribution of your static and dynamic web content, such as.&nbsp;HTML, .css,&nbsp;js, and image files, to your users. CloudFront delivers your content through a worldwide network of data centers called edge locations. I have used CloudFront on multiple projects to ensure any static assets such as anything hosted in S3 can be easily served at the edge to ensure faster response times across multiple regions/availability zones.

AWS CloudFront

Amazon Route 53 is a highly available and scalable Domain Name System (DNS) web service. Route 53&nbsp;connects user requests to internet applications running on AWS. I use Route 53 to manage my hosted zones and domains, including the administration of DNS records for site verification, routing, and email setup.

AWS Route53

Identity and access management (IAM) is&nbsp;the discipline that enables the right individuals to access the right resources at the right times for the right reasons. I use AWS IAM when provisioning my AWS resources both via the AWS GUI and the AWS CLI. I try to employ best practices when it comes to the creation of roles and creation of 'least privilege' 'policies.

AWS IAM

Amazon DocumentDB is a scalable, highly durable, and fully managed database service for operating mission-critical MongoDB workloads. It is fully compatible with MongoDB and is the AWS version of cloud-hosted 'Atlas'. I have mostly used Atlas up until recently when we migrated our atlas clusters over to DocumentDB as part of some company housekeeping to ensure all of our internal systems are housed within AWS and easily reconstructable using TerraForm or CloudFormation.

AWS DocumentDB

Amazon Cognito&nbsp;lets you easily add user sign-up and authentication to your mobile and web apps. Amazon Cognito also enables you to authenticate users through an external identity provider and provides temporary security credentials to access your app's backend resources in AWS or any service behind Amazon API Gateway. I used Cognito to implement authentication &amp; federated login with a hosted login portal for Audiolancer.io. I also used Cognito to pass role credentials through to my application to ensure that certain roles have CRUD permissions locked down to specific S3 Prefixes. After using Cognito for a while, I switched Audiolancer.io to use NextAuth authentication instead because it gave me easier access to my own pool of users via my own database rather than having them managed by AWS which I found to be too much of a limitation.

AWS Cognito

Express JS is a web framework that makes it extremely fast to write web applications and rest APIs with Node without having to write often repeated and mundane code. If I am writing a web server or API and the project does not require NextJS or NuxtJS then I will always opt to use ExpressJS purely because it is faster, more efficient, and easier to manage than writing pure NodeJS code.

Express JS

NodeJS is a javascript runtime environment that allows you to run a server using javascript. I have been using NodeJS since 2018 and it is my primary language for any backend tasks. I have used it with ExpressJS for web servers, bots, and data manipulation tasks. In recent years, when building web applications in the form of traditional websites, I have moved away from standalone Node projects in favour of NextJS.

Node JS

cPanel is&nbsp;an online Linux-based graphical interface (GUI) used as a control panel to simplify website and server management. cPanel allows you to publish websites, manage domains, organize web files, create email accounts, and more. I have used cPanel as a website owner and hosting reseller to set up clients with websites. (predominantly WordPress)

Cpanel

Amazon SQS is a message queue service used by distributed applications&nbsp;to exchange messages through a polling model and can be used to decouple sending and receiving components. Amazon SQS is used by a developer&nbsp;to reduce the complexity of managing and operating a service. Using this we can store, send, receive and manage messages between two services Instead of sending messages directly we can send them through the intermediate queue. I use SQS throughout my other AWS service implementations for sending data between components to improve reliability and prevent any messages or events from being dropped in the unlikely event of any downtime. I also use SQS for dead letter queues for any failed events to be automatically retried as well as for logging purposes.

AWS SQS

Amazon Elastic Compute Cloud&nbsp;(Amazon EC2) is a compute platform that provides a wide selection of server instance types optimized to fit different use cases. Instance types comprise varying combinations of CPU, memory, storage, and networking capacity and give you the flexibility to choose the appropriate mix of resources for your applications. I have used AWS EC2 to host Apache Servers &amp; Machine Learning docker containers with TensorFlow as well as running bots for forex trading, Instagram, and Tinder.

AWS EC2

The AWS Serverless Application Model (AWS SAM) is&nbsp;an open-source framework for building serverless applications. It provides shorthand syntax to express functions, APIs, databases, and event source mappings. With just a few lines per resource, you can define the application you want and model it using YAML. I use AWS SAM for simplifying the development and deployment process of Lambda functions and layers on projects such as Audiolancer.io &amp; Picks &amp; Parlays.

AWS SAM

AWS CloudFormation is a service that&nbsp;gives developers and businesses an easy way to create a collection of related AWS and third-party resources, and provision and manage them in an orderly and predictable fashion. I use AWS CloudFormation for provisioning &amp; versioning AWS infrastructure as code (IAC) for projects such as Audiolancer.io &amp; Picks &amp; Parlays. I prefer this approach because having the infrastructure version controlled makes any infrastructure updates much easier to setup, teardown or replicate across environments.

AWS CloudFormation

AWS EventBridge delivers a stream of real-time data from your applications, software as a service (SaaS) applications, and AWS services to targets such as AWS Lambda functions, HTTP invocation endpoints using API destinations, or event buses in other AWS accounts. I use EventBridge on sites such as Audiolancer.io and Picks&amp;Parlays for the scheduling &amp; execution of Cron Jobs with the use of Lambda functions that operate directly on my applications database layer.

AWS EventBridge

WebPack is module bundler. <ul><li>It helps to use different javascript files without having tension that will load first.</li><li>It makes code shorter.</li><li>It helps in converting many files other than javascript into modules.</li><li>It compiles different javascript modules</li></ul> I have been using WebPack for compilation tasks such as minification, compression, and compiling sass since 2017. It is used behind the scenes as a compiler in many other projects but rarely requires any changes outside of the initial site configuration.

Webpack

I have been both consuming and building REST APIs since learning Javascript and NodeJS. I have built REST APIs with Node &amp; Express, Next, Nuxt, PHP, and AWS API Gateway and lambda functions. I have been testing, documenting APIs, and code versioning them with tooling such as Cypress &amp; Postman. I have found by doing this it is much easier to create API definitions and follow OpenAPI 3.1 specifications.

Rest APIs

npm is&nbsp;the world's largest Software Registry. The registry contains over 800,000 code packages. Open-source developers use npm to share software. I have been using npm extensively since I first started using Javascript modules in 2016. More recently I have started using Yarn over npm in some Vue projects because it tends to be more reliable in terms of information logging and checking of peer dependencies.

Bootstrap is&nbsp;an open-source front-end development framework for the creation of websites and web apps. Designed to enable responsive development of mobile-first websites, Bootstrap provides a collection of syntax for template designs. I first started using bootstrap as a WordPress/PHP developer with the lamp stack. Whilst I rarely use Bootstrap now, I do occasionally use packages such as 'react-bootstrap' within react projects to build and prototype frontends faster than writing pure CSS or Sass.

Bootstrap

The Flexible Box Module, usually referred to as flexbox, was designed as a one-dimensional layout model, and as&nbsp;a method that could offer space distribution between items in an interface and powerful alignment capabilities. Whilst not really a skill by itself, I have mentioned it here as it tends to be a feature that some developers struggle to implement correctly. I learned this integral element of everyday CSS to assist with building frontend mobile optimized WordPress templates using tools such as FlexBox froggy to easily visualize how Flex properties affect the layout of a page. I also spent some time learning CSS grid mode however I rarely find myself needing to use this due to making use of CSS framework components such as Material UI.

Flexbox

jQuery is the most popular library on the web today. It's a library of JavaScript functions that make it easy for webpage developers to do common tasks like&nbsp;manipulating a webpage, responding to user events, getting data from servers, building effects and animations, and much more. I have used jQuery across many WordPress personal and client websites that I have managed as part of my music industry hosting platform as well as across all WordPress projects when I was a junior web developer prior to moving towards a headless WordPress architecture and react. JQuery is a great tool for headed WordPress systems but it is no longer something that I work with very often due to seeing it as a legacy language that is now being replaced by JS frameworks such as React &amp; Vue.

jQuery

MySQL is&nbsp;a database management system. It may be anything from a simple shopping list to a picture gallery or the vast amounts of information in a corporate network. To add, access, and process data stored in a computer database, you need a database management system such as MySQL Server. I have worked with MySql databases a lot, specifically with all aspects of WordPress, although I have also done a Javascript App with MySql instead of my usual MongoDB solution. From standard DB queries and operations to full site migrations. I usually use tools such as PHPmyAdmin &amp; MySql Workbench. I rarely work with MySql databases now as MongoDB is my database of choice for new projects due to its NoSQL architecture and ease of integration with JS projects.

MySql

Javascript is used by programmers across the world&nbsp;to create dynamic and interactive web content like applications and browsers. JavaScript is so popular that it's the most used programming language in the world, used as a client-side programming language by 97.0% of all websites. It is also my main language of choice in its many different forms. I Learned JS through sources such as MDN, W3, and a lot of Udemy courses, and further developed my knowledge through extensive learning &amp; practice code-labs, as well as applying it within my day-to-day work.

Javascript

Hypertext Markup Language, or HTML, is&nbsp;a programming language used to describe the structure of information on a webpage. Together, HTML, CSS, and JavaScript make up the essential building blocks of websites worldwide, with CSS controlling a page's appearance and JavaScript programming its functionality. HTML was the first thing I learned as a self-taught web developer using free resources on MDN &amp; the W3 Schools website.

Html

CSS&nbsp;(Cascading Style Sheets) allows you to create great-looking web pages. I Learned CSS via articles from MDN, W3, and Css-tricks.com as well as various Udemy courses. Since learning CSS, I have moved to CSS frameworks such as Bootstrap, Vuetify, Sass, Material UI, Reactstrap &amp; TailwindCSS.

WordPress is&nbsp;a content management system (CMS) that allows you to host and build websites. WordPress contains plugin architecture and a template system, so you can customize any website to fit your business, blog, portfolio, or online store. WordPress was my introduction to web development and I have developed and maintained hundreds of WordPress sites, and created and maintained custom themes and plugins mostly for internal use within internal business use. After a year or so of traditional (headed) WordPress site development, I moved to develop headless WordPress sites, utilizing React as a front-end framework. I occasionally maintain WordPress sites using the LAMP stack, however when given the choice I prefer to always work with full javascript applications utilizing the MERN or MEVN stack due to preferring frameworks such as Next &amp; Nuxt and the ease of writing entire systems entirely in javascript. Aside from working on a multitude of sites in the gaming and gambling sector, I have also built various WordPress sites that utilize Woocommerce, backend storage with AWS, booking systems, and client portals for hosting websites for selling courses such as my music production website www.studioslave.com.

WordPress

PHP is an open-source server-side scripting language that many devs use for&nbsp;web development. It is also a general-purpose language that you can use to make lots of projects, including Graphical User Interfaces (GUIs). PHP is the backend language of the internet prior to the wider adoption of Javascript and other languages as servers. I learned PHP whilst administrating WordPress sites and it was my predominant backend language for a few years. I have since moved away from actively developing PHP projects in favour of fully Javascript applications however I do still occasionally maintain PHP and WordPress sites.

Josh Van Riel