Quoting Liz Fong-Jones
Explores how AI language models shift a programmer's role from writing code to managing context and providing detailed specifications.
SimonWillison.net is the long-running blog of Simon Willison, a software engineer, open-source creator, and co-author of the original Django framework. He writes about Python, Django, Datasette, AI tooling, prompt engineering, search, databases, APIs, data journalism, and practical software architecture. The blog includes detailed notes from experiments, conference talks, and real projects. Readers will find clear explanations of topics such as LLM workflows, SQL patterns, data publishing, scraping, deployment, caching, and modern developer tooling. Simon also publishes frequent micro-posts and TIL entries that document small discoveries and tricks from day-to-day engineering work. The tone is practical and research oriented, making the site a valuable resource for anyone interested in serious engineering and open data.
133 articles from this blog
Explores how AI language models shift a programmer's role from writing code to managing context and providing detailed specifications.
shot-scraper 1.9 CLI tool released, featuring a new -x option to extract page resources and accessibility command fixes.
D. Richard Hipp discusses SQLite's rigorous aviation-grade testing methodology, contrasting it with PostgreSQL's peer-review approach.
The core challenge of programming is translating ambiguous human thought into precise computational logic, not just writing code.
Explains SQLite's selective contribution process and the required public domain copyright release, with a query showing contributor stats.
Explores how AI will expand knowledge work by making tasks cheaper, leading to new projects and activities we don't do today.
A developer creates a scraper to track the latest versions of GitHub Actions, helping coding agents avoid stale workflows.
Author details how Substack's content filter blocked a newsletter containing a SQL injection exploit example, citing a 'Network error'.
Boris Cherny shares his experience using Claude Code + Opus 4.5 to write all code for 259 PRs in a month, highlighting AI's coding progress.
A review of textarea.my, a minimalist browser-based text editor that stores data in the URL hash, highlighting its clever JavaScript techniques.
An analysis of the performance optimizations in uv, a fast Python package installer, focusing on HTTP range requests and compact version representation.
Rob Pike's angry reaction to receiving an AI-generated 'thank you' email from the 'AI Village' project, sparking debate about AI ethics and spam.
Introduces claude-code-transcripts, a Python CLI tool for converting Claude Code sessions into detailed, shareable HTML transcripts.
A demonstration of different uv init command options for Python project setup, with examples hosted on GitHub.
Redis creator Salvatore Sanfilippo discusses why he would have chosen JavaScript over Lua for Redis scripting if MicroQuickJS existed in 2010.
Explores MicroQuickJS, a tiny JavaScript engine for embedded systems, as a potential sandbox for running untrusted code with strict resource limits.
A developer uses the Claude in Chrome browser agent to navigate the Cloudflare dashboard and solve a CORS configuration problem.
A visual essay explaining LLM internals like tokenization, embeddings, and transformer architecture in an accessible way.
OpenAI releases GPT-5.2-Codex, a model optimized for agentic coding with improvements in refactoring, Windows performance, and cybersecurity.
Anthropic's Agent Skills specification becomes an open standard, detailing its lightweight design and current industry adoption.