Tom Shaw

Running my coding agents completely offline, and getting the work done quickly rather than waiting hours for the task to finish. I spend days of my life on trains or planes without wifi on a monthly basis, so this is a game-changer

Our feeds are full of opinions about how data centers operate. It's not surprising when most of the infrastructure powering the internet, and the AI boom, operates completely behind the scenes. Over the past few months, I’ve had the opportunity to get behind the scenes of that infrastructure with @MetaNewsroom . For my latest article, I wanted to dig into 3 of the biggest claims I hear about data centers. ⬇️ about.fb.com/news/2026/10/de…

Data center liquid cooling is a hot topic! 📷 But how does it actually work? 📷 Over the past few months I've made multiple visits behind the scenes at some of @Meta infrastructure labs and data centres. Yesterday my own written article was published to @MetaNewsroom , diving into my learnings and explaining how closed-loop liquid cooling systems deployed at Meta’s AI data centres actually work. about.fb.com/news/2026/08/cl…

You can now integrate @github Copilot directly into your own programming projects! 🤖 The Copilot SDK allows you to quickly interact with the GitHub Copilot CLI installed on any device. Making it much easier to build dev tools, internal tools, and other fun projects for GitHub Copilot users! 💻 I’m using it to build a new version of my Quotation Agent, allowing me to instantly price up my projects for clients from the command line! 💰 (Paid Partnership with GitHub)

3 NEW Features added to F1 Race Replay! 🏎️ The new features include: - Driver Tyre Information 🛞 - Driver Telemetry Insights 📈 - Track Status Indicators 🚥 We’ve also already had the first external pull request, where another developer added in the ability to resize the window to fit any screen! 🖥️ It’s open-source, so anyone can download the code, have a play and even contribute features of their own for everyone else to enjoy! 🚀

My latest micro project “𝗙𝟭 𝗥𝗮𝗰𝗲 𝗥𝗲𝗽𝗹𝗮𝘆” allows you to watch the race from above using telemetry from the FastF1 python package! 🏎️ Tech Stack 💻 - Python 🐍 - FastF1 (data source) 📈 - Arcade (game library used to create viewer) 📺 Maybe it's not the most fun way to watch the sport, but it's a pretty cool project that any aspiring builder and F1 fan can build in a day! 🚀

Okay, I think I’m living the dream programmer / content creator life 🙌

At last week’s AWS Summit, I stumbled across this chessboard that was controlled by a VR game. Using Unreal engine and a Bluetooth chessboard, the developers were able to create a game that synced the state of the game from the VR headsets to a real-world chessboard. My understanding is: you move a piece in the VR world, and the piece moves by itself on the real-world chessboard. The applications of technology like this can go far beyond a chessboard. Think about typically “in-person” tasks such as medical surgeries. These technologies still feel like a long way away, but probably aren’t as far away as we think,. Oh, and all of the game assets (characters, chess pieces, etc) were generated using AI.

Here’s how @netflix serves content to you in milliseconds! 🎞️ Using Open Connect (their Content Delivery Network), they’re able to locally cache content in different areas all around the world based on global show preferences. Their microservices architecture enables them to distribute workloads across hundreds of server instances, instead of being restricted by vertical scaling limitations.

If you're a programmer that loves Formula 1, you need to check out the FastF1 python package! 🏎️ And no, it's not deprecated like people thought. ‼️ To clear up any confusion, FastF1 previously used the Ergast API as a data source, which is now deprecated. FastF1 has since moved over the Ergast's successor "Jolopica".

I've just bought 6 Raspberry Pis. +10 points to anyone who can guess what they're for 👀

This is probably my favourite developer tool at the moment! 💻 AWS Infrastructure Composer enables you to visually design your AWS projects, instead of using YAML. 🚀 Having the ability to visualise your infrastructure is game-changing for spotting inefficiencies and building better projects faster. It saves inside of your repository, so you can open and update your infrastructure design at any time during your project.

Every AI tool needs data! 🤖 So when I wanted to build a meeting’s agent for my AI assistant project, I needed to figure out how I was going to capture what was said in my meetings, and feed it to the AI assistant. 🧠 Using some serverless AWS services, I’ve created this simple data pipeline that retrieves the information from a meeting and uploads it to my knowledge base as soon as the meeting has ended! 💻 So instead of taking notes myself, I can just ask my AI assistant! 💬

Just landed in Helsinki, Finland! 🇫🇮 Why am I here? I'm going to be checking out some pretty cool tech tomorrow. 💻 I can't say what just yet, but it's definitely interesting enough that I've spent 3 hours on a plane to see it in person! ✈️

Here’s how I’m using Python to automate my life! 🐍 Let me introduce you to my new project. A team of AI Agents to help me run my business. 🤖 I’ve found myself doing the boring tasks over the past few months, so I’m building this project to automate those tasks, so that I can focus my time on the things that I love and that drive the most value to the business. 🚀 If you’d like to stick around for the journey, follow for more and turn notifications on! 🔔

I had so much fun shooting this video on why I use PyCharm with with @jetbrains 🧠

Should you be learning to code in 2025? 🤔 Coding isn't just about writing programs. It transforms how you approach challenges and solve complex problems. And yes, you get to build some pretty amazing things along the way!

And just like that, I've stopped myself from spending all of my OpenAI Credits like I did last night

What are everyone’s thoughts on @crewAIInc ? 🤖 I’ve been playing around with it recently and I’m undecided 👀