Drag to turn the sky · scroll or pinch to move · choose any star to go there
Turning on the stars…
Topic · AI
Working with LLMs
79 links saved from 2023 to 2026; 61 with a note, 61 in a Weekly Thing.
- Most from
- simonwillison.net 9, github.com 7, microsoft.com 3, allenpike.com 2, marginalrevolution.com 2
2026
- The rise and fall of Jev
- MicroLLM lab — tiny LLMs, Q4, in your browser
- LLM Attention Visualization
- My local LLM was slow and unfocused until I added this one missing piece
- Constraint decay: The Fragility of LLM Agents in Backend Code Generation
- forge: A Python framework for self-hosted LLM tool-calling and multi-step agentic workflows
- LLMs are functions, not brains.
- Build with Micro.blog
- 2026 is the year of self-hosting
- LLM problems observed in humans
2025
- 2025 LLM Year in Review
- Context plumbing (Interconnected)
- onyx: AI Chat with advanced features that works with every LLM
- llm-council: LLM Council works together to answer your hardest questions
- The Illusion of Thought: Chain of Thought Lies
- Three Years from GPT-3 to Gemini 3
- 2 Years of ML vs. 1 Month of Prompting
- toon: 🎒 Token-Oriented Object Notation – JSON for LLMs at half the token cost
- LLMs As Conduits for Data Portability Between Apps
- Building Etsy Buyer Profiles with LLMs | Etsy Engineering
- Some thoughts on LLMs and Software Development
- Why LLMs Can't Really Build Software
- LLM Embeddings Explained: A Visual and Intuitive Guide
- Coding with LLMs in the summer of 2025 (an update)
- The Big LLM Architecture Comparison
- Secure & reliable LLMs | Promptfoo
- Why I'm Dialing Back My LLM Usage
- The Conversational SDLC | Joseph Bironas
- Design Patterns for Securing LLM Agents against Prompt Injections
- God is hungry for Context: First thoughts on o3 pro
- My AI Skeptic Friends Are All Nuts
- LLMs are weird, man
- Vibe coding for teams, thoughts to date | Kellan Elliott-McCrea: Blog
- AI-assisted development needs automated tests
- prompt-eng-interactive-tutorial: Anthropic's Interactive Prompt Engineering Tutorial
- Large Language Models can run tools in your terminal with LLM 0.26
- GPT-4.1 Prompting Guide | OpenAI Cookbook
- Ditching Obsidian and building my own
- Embeddings are underrated
- llm-prices.com
- As an Experienced LLM User, I Actually Don't Use Generative LLMs Often
- Post-Chat UI
- Using LLMs as the first line of support in Open Source
- Introducing KBLaM: Bringing plug-and-play external knowledge to LLMs
- The Resistor Network: Thinking Different, Thinking Slowly: LLMs on a PowerPC Mac
- How Much Are LLMs Actually Boosting Real-World Programmer Productivity?
- Here’s how I use LLMs to help me write code
- 🔭 The Einstein AI model
- The Differences between Deep Research, Deep Research, and Deep Research
- Hallucinations in code are the least dangerous form of LLM mistakes
- My LLM codegen workflow atm
- Run LLMs on macOS using llm-mlx and Apple’s MLX framework
2024
- December in LLMs has been a lot
- TinyTroupe: LLM-powered multiagent persona simulation for imagination enhancement and business insights.
- Everything I've learned so far about running local LLMs
- How we saved hundreds of engineering hours by writing tests with LLMs
- LLM Visualization
- What We Learned from a Year of Building with LLMs (Part I)
- Mapping the Mind of a Large Language Model Anthropic
- Anthropic — Prompt library
- Notes on how to use LLMs in your product.
- GraphRAG: Unlocking LLM discovery on narrative private data
- The pain points of building a copilot - Austin Z. Henley
2023
- Phi-2: The surprising power of small language models
- Learnings from fine-tuning LLM on my Telegram messages
- LLM can now be installed directly from Homebrew
- 5 learnings from classifying 500k customer messages with LLMs vs traditional ML
- Going Way Beyond ChatGPT
- privateGPT: Interact privately with your documents using the power of GPT
- Prompts for Work & Play: Launching the Wolfram Prompt Repository—Stephen Wolfram Writings
- Open-Source LLMs
- All the Hard Stuff Nobody Talks About when Building Products with LLMs
- Inside GitHub: Working with the LLMs behind GitHub Copilot
- LLMs and neurodiversity
- An example of LLM prompting for programming
- Eight Things to Know about LLMS
- Free Dolly: Introducing the World's First Open and Commercially Viable Instruction-Tuned LLM
- Replacing my best friends with an LLM trained on 500,000 group chat messages
- The AI singularity is here