👋 Welcome to my website!

Like many people, I've been using tools like ChatGPT for some time now and they're great, however, they lock you down to a third party service that you need to pay monthly for to get a good service and they aren't as fun given that you can't really customise them. With this project, I'm going to build my own using Cloudflare AI. It will be capable of full conversations alongside being able to query for external data.

I do quite a few things in Python, in particular, it's a great language for data projects. Recently Cloudflare added beta support for Python in their Workers platform, so I thought I'd investigate, but for that, I needed a cool project to work on, so not only is this a post about exploring the support for Python in Cloudflare Workers, but it's also about building a chatbot that can interact with eBooks, if Cloudflare Workers with Python actually works anyway...

I came across a new page from Daniel Roe where he sets out his guidelines for his use of AI, I thought I'd do the same.

AI is a big topic in the industry lately, so much so that we end up talking about it in every other meeting, from the latest advancements to the ethical implications, there's a lot to consider. Recently, I've been thinking about my own thoughts and principles around the use of AI, and I thought I'd share them with you.

The "llm-rss-vectorise-agent" project explores the power of large language models (LLMs) combined with Cloudflare’s edge infrastructure to analyze, vectorize, and process RSS feed content. This post takes you through my experience of using it for this project and how I think AI works in this context.

Creating an A/B testing platform from scratch was a unique challenge. I wanted to build a tool capable of running both Bayesian and frequentist experiments, that provided a decent foundation to build more from, but a bit scrappy to keep it as a side project. This post walks through the development journey.

Creating a GraphQL service to retrieve bookmarks from DynamoDB

I participate in a range of projects at the BBC. Most recently, I transitioned the BBC’s Open Source site to the WebCore service to enhance performance, maintainability, accessibility, and user experience. I also contribute to open-source projects, such as SQS Consumer, and help teams improve their experimentation capabilities.

At Accrosoft, I developed a custom CMS called AccroPress, a fully headless system with a React frontend and a Node.js/Express backend. Designed as a hosted service, it was used by various school and job sites worldwide. AccroPress also offered several microservices, including image processing, contact forms, and more.

While at Vacancy Filler, I collaborated with clients like Bouygues Construction UK to launch career site initiatives that attract and inform potential candidates about roles across their organization. I also worked with brands such as the Woodland Trust, Longleat, and the Devonshire Group, among others.