Joe Brown

I led a team of developers in building a greenfield cross-platform mobile application using React, Tailwind, Ionic, and Capacitor. I designed and implemented the core architecture to ensure long-term maintainability, introduced Playwright E2E testing, and established CI/CD pipelines with GitHub Actions.

I also led the development of an event-driven microservice that ingested 23 million IoT events/day and calculated energy usage in real time. This service was built on AWS, using Kinesis, Firehose, DynamoDB, S3, Athena, Glue, Lambda and API Gateway.

I also worked on our new SSO service, improving security by implementing email-based MFA, adding Azure SAML login for staff and added passwordless login for customers. I also built functionality to seamlessly migrate "legacy" accounts from our web portal to SSO. This service is built on Cloudflare, utilising Workers, Pages, D1, KV, AWS Cognito for authentication, and React + Tailwind on the frontend.

Working across teams, I contributed to numerous projects, including creating a framework-agnostic frontend library to visualise hardware topologies and interface with our new HEMS (Home Energy Management System).

As part of broader infrastructure improvements, I was involved in migrating our monorepo from npm to pnpm. I rewrote several GitHub Actions, notably the project detection step, which improved from 4 minutes to 10 seconds.

In addition to development work, I mentored multiple junior developers, teaching core software engineering principles, and helped Flutter developers skill up in React and TypeScript.

I also conducted technical interviews for a junior full-stack developer role and a web developer role.

I worked as a frontend developer on BrewMan, one of the UK's leading brewery and distillery management solutions.

While my primary focus was on frontend development using Vue, Vuetify, and Sass, my role expanded over time to include backend development with C# and PostgreSQL.

• Rebuilt our corporate website from the ground up using Nuxt, improving SEO and maintainability.
• Developed a scalable, TypeScript-based UI system to enhance code reuse and type safety across the most commonly used areas of BrewMan.
• Migrated the proprietary authentication system to Firebase and built admin tools in Node.js.
• Optimised frontend performance as the business scaled, refactoring and improving key areas of the codebase.
• Championed Vue 3 migration - I migrated a significant amount of our codebase's Vue 2 Options API and class components to the Composition API, ensuring a smooth transition while maintaining stability and backwards compatibility.
• Implemented a customisable dashboard, both frontend and backend, which became the new homepage of BrewMan.
• Mentored a new mid-level developer, helping them onboard and gain familiarity with the codebase and tech stack.
• Played a key role in migrating BrewMan's most widely-used component, our bespoke grid, to AG Grid.
• Implemented dark mode support, refactoring CSS across the entire application to ensure seamless theme switching.
• Championed Cypress E2E testing, improving test coverage and reliability.
• Enhanced our Progressive Web App used by delivery drivers, introducing new features and performance optimisations.

Key modules studied:

• Software Quality and Testing
• Advanced Development
• Deep Learning
• Networks and Cyber Security
• Systems Design
• Data and Databases

For my Final Year Project, I developed a web-based CRM application for sales representatives in the brewery and distillery industry, designed to integrate with BrewMan. The project was built based on requirements gathered from the sales team at Premier Systems.

Bournemouth University selected my dissertation as a reference for future final-year students.

For my 2nd year computer science project, I built a maze-solving CLI application in VB.NET that implemented both Dijkstra's algorithm and the A* pathfinding algorithm.

This application parsed an image of a maze, with a start and end point, and output the same image overlaid with the shortest path between the two points. It also rendered the path finding process in the console, step by step, animated in real time.