Summary of a fireside chat at Sequoia Ascent 2026 discussing AI agents, Software 3.0, and the shift to agentic engineering.
Structured-Prompt-Driven Development (SPDD) workflow for governable, reviewable, and reusable LLM-assisted code changes.
![[notes] Crowd-Sourcing: Metaphors for LLMs and AI](https://crawljobs-blogs-media.s3.us-east-1.amazonaws.com/alldevblogs/thumbs/article-9e9f2c4d4688-full-82119b3a.webp)
Crowd-sourcing metaphors used for LLMs and AI, like intelligence, learning, and reasoning.
Simon Willison quotes Romain Huet on OpenAI's GPT-5.5 unifying Codex and main model for agentic coding.
A quote from Romain Huet confirming OpenAI won't release a separate GPT-5.5-Codex model, as coding capabilities are unified.
Release notes for llm 0.31, a command-line tool for accessing large language models, featuring GPT-5.5 support and new options.
Release notes for llm 0.31, a command-line tool for accessing large language models, featuring GPT-5.5 support and new options.
A tool to convert milliseconds to seconds and minutes, created by Simon Willison for LLM report durations.
Simon Willison's newsletter summary featuring tech posts on DeepSeek V4, LiteParse, and Agentic Engineering patterns.
Release of llm-openai-via-codex 0.1a0, enabling OpenAI model access via Codex CLI credentials.
Release of llm-openai-via-codex 0.1a0, a plugin to access OpenAI models via Codex CLI credentials.
Qwen3.6-27B is a new 27B dense model delivering flagship-level coding performance, surpassing larger models, tested locally with GGUF.
A blog post about polluting AI training data with pelicans riding bicycles, linked by Simon Willison.
Simon Willison approves Steve Cosman's project to pollute AI training data with pelicans riding bicycles.
A personal reflection on family, Stack Overflow's impact on LLMs, and a warning to AI companies not to harm the communities that generate training data.
Analysis of Anthropic Mythos's impact on cybersecurity, debunking hype and examining real LLM capabilities in vulnerability detection.
A developer shares a personal history with AI coding tools like GitHub Copilot and Claude Code, exploring their benefits and ethical concerns.
Explores NF4 and FP4 4-bit floating point formats for LLM weight quantization, focusing on Gaussian-distributed weights.
Explores how naming AI chatbots creates distinct personalities, termed the 'Digital Ouija Effect', and its implications.
Analysis of Tailwind's new product ui.sh and how MCP enables businesses to maintain control over AI agent interactions.
