domain | simonwillison.net |
summary | The content appears to be a series of blog posts or updates from various sources discussing advancements in artificial intelligence (AI), particularly focusing on language models like ChatGPT and Claude. The posts cover topics such as:
1. Browserbase's CAPTCHA-solving capability, demonstrated during a Gemini demo. 2. NVIDIA DGX Spark hardware review and the early stages of its ecosystem. 3. Claude's ability to create complete Datasette plugins. 4. Vibe engineering, implying AI development or related work. 5. OpenAI DevDay 2025 live blog, possibly discussing industry events or AI news. 6. Embracing a parallel coding agent lifestyle, suggesting new ways of working with AI models. 7. Designing agentic loops, which might refer to creating closed systems for AI agents. 8. Claude Sonnet 4.5 being hailed as potentially the best coding model currently available. 9. Discussion on defining "agent" in AI, suggesting a more standardized term. 10. Reviews of Claude's Code Interpreter and its confusing naming convention. 11. Recreating Apollo AI adoption charts using GPT-5, Python, and Pyodide. 12. GPT-5's impressive search capabilities, dubbed "Thinking in ChatGPT." 13. Exploration of historical events or figures like the East Storehouse and Operation Mincemeat. 14. Inconsistencies in performance across open-weight large language models (LLMs). 15. GPT-5's key characteristics, pricing, and model card details. 16. Introduction to OpenAI's new Apache 2 models, notably good performance. 17. ChatGPT users' user-agent investigation. 18. Challenges in designing privacy preferences for sharing dialogs with ChatGPT. 19. Reverse engineering updates to Claude using LM Studio and Open WebUI. 20. Utilizing AI for game development, such as writing a Space Invaders clone using GLM-4.5 Air and MLX. 21. Reverse engineering GitHub Spark. 22. Creating a schedule app (vibe) entirely on a phone using vibe scraping and coding. 23. Discussing Django's origins at its 10th anniversary event. 24. Grok search tool for finding information related to specific entities, e.g., X from:elonmusk. 25. Phoenix.new as an entry into prompt-driven app development. 26. LLMs' evolution illustrated by pelicans on bicycles (a creative metaphor). 27. Tips for prompting ChatGPT, such as when addressing UK technology secretary Peter Kyle. 28. Recreating Theos SnitchBench with LLM for evaluating language models. 29. Talking about AI and jobs with Natasha Zouves for News Nation. |
title | Simon Willison’s Weblog |
description | Simon Willison’s Weblog |
keywords | code, python, agents, model, coding, llms, using, like, have, haiku, context, tokens, skills, version, first, more, prompt |
upstreams |
blogroll.org, adactio.com, glazkov.com, tedunderwood.com, walknotes.com, waxy.org |
downstreams |
npmjs.com, nytimes.com, youtube.com, wired.com, mastodon.social, toot.cafe, daringfireball.net, mozilla.org, alexwlchan.net, github.com, twitter.com, observablehq.com, huggingface.co, bsky.app, mlc.ai, openai.com, menlovc.com, ollama.com, datasette.io, wikipedia.org, zachholman.com, htmx.org, ycombinator.com, microsoft.com, google.com, v0.dev, vercel.com, fosstodon.org, modelcontextprotocol.io, claude.ai, sourcegraph.com, zed.dev, python.org, semgrep.dev, jonathanadly.com, colivara.com, arxiv.org, jvns.ca, esm.sh, aider.chat, dustycloud.org, social.coop, w3.org |
nslookup | A 172.67.136.172, A 104.21.56.206 |
created | 2024-02-14 |
updated | 2025-10-17 |
summarized | 2025-10-17 |
|
|