LIFTLOG started as a gym-tracking app built around my own training, then grew into a full fitness platform covering strength training, cardio (walking, running and cycling) and nutrition planning, with an AI coaching layer on top. From the outset I planned it as a freemium product with a subscription model.

Approach & design — I ran LIFTLOG as an Agile project, breaking the work into small iterations and prioritising features release by release. Before writing any code, I designed the key screens and user flows in Figma, working from wireframes through to high-fidelity designs. Those designs then fed into a token-based design system in code: one source of truth for colour, typography and spacing across every screen, with dark mode supported from day one rather than bolted on later. This kept the app visually consistent as new features were added each iteration.

Figma designs: Home, Active Workout and Account screens

Development — With the designs in place, I built the app with React Native, Expo and TypeScript, with Firebase handling authentication and data, and Cloudflare Workers AI powering the coaching and voice-transcription features. Working in iterations, I delivered it feature by feature: workout logging with drop sets, per-limb tracking and supersets; a trainer-client system; full internationalisation across 14 languages; a GPS walk, run and bike tracker with live mapping and repeatable routes; and a menstrual cycle tracker. Each one was layered onto the existing architecture without needing a rebuild.

The built Home screen running in the Xcode iOS Simulator

Testing & launch prep — Testing started in Expo Go and the browser, until native modules (maps and in-app purchases) meant moving to a Mac with Xcode and the iOS Simulator. That surfaced real issues I worked through, from Firestore security rules to platform-specific crashes and Firebase Auth API changes. I integrated RevenueCat for subscriptions and AdMob for ads, and tested both end-to-end on sandbox infrastructure.

LIFTLOG is now in final pre-launch preparation. Apple Developer Program enrolment is complete, with App Store Connect setup, legal documentation and translation work remaining before public release.

Tech stack: React Native, Expo, TypeScript, Firebase (Auth & Firestore), Cloudflare Workers AI, RevenueCat, AdMob, Xcode, Figma

The brief — Joe Browns wanted to grow SMS as a marketing channel alongside email. That meant capturing customers' mobile numbers and consent on the website, and connecting them to Bloomreach so marketing could send automated, personalised messages.

Stakeholders — CRM, Marketing, Digital Design and Product internally, plus Bloomreach as the external platform partner. Each had different priorities, from the customer journey and brand look to data capture and campaign automation.

How I ran it — I owned delivery end to end. I gathered requirements with CRM and Marketing, turned them into scoped tasks in Jira and planned them into sprints, and acted as the link between the internal teams and Bloomreach so decisions didn't stall between them. On the build side, I created the custom attributes and workflows that take a customer from entering their number on site through to being signed up for SMS. I then coordinated testing across the full journey before a controlled rollout.



Outcome — SMS sign-up went live as a fully automated journey, giving marketing a new owned channel and improving customer engagement workflows without adding manual work for the CRM team.

Tools: Jira, Magento 2, Bloomreach, HTML, CSS, JavaScript

Huzis Hotels is a personal front-end project: a Dubai travel site inspired by a holiday I had coming up. I set the brief myself, treating it like a client project with a clear goal: a site that feels visually rich and interactive, but stays simple for visitors to navigate and explore Dubai's attractions, culture and activities.

Build — I used HTML, CSS and JavaScript with Bootstrap for a fully responsive layout across devices. Interactive features were built with focused libraries rather than one heavy framework: jQuery and Owl Carousel for scrolling landmark galleries, Classie for CSS animation triggers, jQuery countTo for animated Dubai fact counters, and Superfish for a multi-level dropdown menu.

Why it matters — The project was a chance to push newer layout techniques and more advanced styling while balancing visual ambition against usability, the same trade-off I now manage on live e-commerce projects.

See the website here: Huzis Hotels

The problem — The Joe Browns website used a simple drop-down menu, which made it slow for customers to reach specific product ranges in a large catalogue.

My role — After researching navigation patterns, I pitched a redesign to my senior developer and took ownership of delivering it. I planned the work through the SDLC, from research and requirements through design, build and testing, and set myself a realistic timeline that I delivered against. I brought my senior in at key points for review and sign-off.

The result — The new megamenu adds category images, icons and a deeper multi-level sub-menu, so customers can get to the exact products they want faster and with fewer clicks.

See the menu here: Joe Browns New Navigation Menu

The brief — As part of an SEO initiative, management wanted more interactive, engaging content on the Joe Browns website. After reviewing the site, I proposed an interactive timeline on the About Us page to tell the brand's story.

Scope & stakeholders — What started as a small idea grew into a cross-team project. I gathered input from the wider e-commerce team, worked with Marketing on the content, and brought in the owner, Simon Brown, to make sure the story was accurate, from how Joe Browns started to where it is today. Coordinating that input and keeping the content aligned was as much a part of the project as the build itself.

Delivery — I built the timeline in Magento 2, so the brand's history became an interactive feature rather than a block of text, supporting the SEO goal with richer on-page content.

See the timeline here: Joe Browns About Us Timeline

The brief — A deliberately small specification for a company directory: a CRUD application for managing staff records. I treated the spec as a baseline, meeting the core requirements and then scoping extra usability improvements for the end user.

Approach — I ran it through the software development life cycle. I started by planning the stack and technical approach, then wireframed the interface, then built to those designs.

Build — A full-stack application using PHP and MySQL on the back end, with AJAX, jQuery, JavaScript, HTML and CSS on the front end, so records can be created, viewed, edited and deleted without page reloads.

See the company directory here: Company Directory

A mobile-first responsive online application that can also be used on a desktop that profiles all nations by presenting geographical, economic, and political statistics about the chosen country. This is aided primarily by the widespread use of APIs. When you click on a country on the map, a modal displays with pertinent information such as the country's name, currency, weather forecasts, and data.

The problem — Doozy Vape Co had an established WooCommerce site on WordPress, but traffic had dipped and new products weren't getting the visibility they deserved.

Discovery — I was brought in for my SEO experience. An audit showed the agency that built the site hadn't put any SEO in place at all. The site's success so far had come from word of mouth, social media and the strength of the products themselves, which meant there was significant room to grow.

Delivery — I researched target keywords, prioritised the fixes with the biggest impact, and rolled out optimised titles, meta descriptions, URLs and sitemap structure across the product pages. This took the site's SEO from non-existent to a solid foundation, resulting in a ~25% improvement in organic visibility and search rankings.

The brief — Sweven Plates, a family-run business, needed an e-commerce site. It was my first freelance project, and I delivered it alongside a senior developer.

Approach — We followed the software development life cycle: paper sketches first, then Adobe Photoshop designs for client approval, then the build. The owner chose Shopify so the team could manage the site day to day, and that requirement shaped the whole build.

Scope — Because social media was the business's main sales driver, the project also covered setting up their Instagram and Facebook pages so the site and socials worked together to grow reach and sales.

It was my first experience of managing a real client's priorities rather than just writing code, and it set the pattern for how I approach projects now.

The brief — My final-year dissertation project, also covering the Advanced Front-End Web Development module. As a regular gym-goer, I built a gym web app styled to JD Gyms branding, designed to work equally well at home and on a phone in the gym.

Front end — I used Bootstrap 4's grid for a mobile-responsive layout and its navbar for a collapsible hamburger menu on smaller screens. A JavaScript search and filter lets users find their nearest JD Gym across the UK. jQuery AJAX loads the personal trainers section from external data so users can choose their PT, and a maps API plots every gym location from a JSON file.

Back end — JD Gyms' own member area was run by an external company, so I scoped and built my own. A PHP and MySQL login updates the navigation through the user's session and unlocks a CRUD area where members can create, view, edit and delete their gym workouts. Visitors who aren't logged in can still browse workouts through a simple show and hide toggle.

Looking back, it was an early end-to-end build covering scoping, front end, back end and data, all shaped around what a real gym member would need.