Managing a complex Python repository with a shared core, various plugins, and multiple applications used to be a headache involving brittle path dependencies or complex scripts. In 2025, the uv workspace has emerged as the definitive solution for these "monorepo" architectures.

Early users treated LLMs as one-shot text generators. Users focused on clever wording, examples, and formatting of their prompts to achieve their goals. This became known as “prompt engineering”. Prompt Engineering worked for simple tasks, but broke down as soon as people needed memory, tools, retrieval, or multi-step reasoning.

I don’t usually write about security, but as a frequent user of Claude Code, I felt the recent news of the s1ngularity attack warrants special attention. Just as we all started to rely on AI CLI tools like Claude and Gemini to enhance our work, hackers just exploited them in a real supply-chain attack. It’s the first time we’ve seen this in the wild.

I used to spend my time before classes playing the original Desktop Tower Defense. It had a long loading screen, it lagged when there are too many units, and it was written in Flash. I wanted to make a modern JavaScript version that’s smoother and more optimized, but I wasn’t a web developer… until LLMs came along.

I spent more than 40 hours to build a javascript-based game (which you can play here). I use Python day to day, and operating in a new language and framework in a new AI tool presented a steep learning curve. In this post, I share some learnings.

In the world of modern software development, few libraries have had as transformative an impact as Pydantic. What began as Samuel Colvin’s elegant solution to Python’s data validation challenges has evolved into the foundational layer powering everything from high-performance web APIs to cutting-edge AI applications. Today, Pydantic sits at the center of an ecosystem that has become indispensable for developers building reliable, production-ready systems.