Large Language Models (LLM)¶
Coding assistant¶
- ChatGPT Is An Extra-Ordinary Python Programmer
- Developers with AI assistants need to follow the pair programming model
- Hard Truths About Generative AI for Technology Leaders
- My colleague Julius
- Can LLMs write better code if you keep asking them to "write better code"?
- Large Chainsaw Model
- TabbyML: self-hosted AI coding assistant
- The 2025 AI Engineer Reading List
- LLM code generation workflow
- Aider: AI pair programming in your terminal
- codegen: Python SDK to Interact with Intelligent Code Generation Agents
- I'd rather read the prompt
- The Problem with "Vibe Coding"
- Emerging Patterns in Building GenAI Products
- The Hidden Cost of AI Coding
- AI code is legacy code from day one
- sourcebot: a self-hosted tool that helps you understand your codebase
- AI Is A Floor Raiser Not A Ceiling Raiser
- Design Partner
- In Praise Of Normal Engineers
- Building your own CLI Coding Agent with Pydantic-AI
- Why Your Prompts Don't Belong in Git
- Designing agentic loops
- Why Generative AI Coding Tools and Agents Do Not Work For Me
- Micromanaged Driven Development: Build all your code with AI and keep full control
- Agentic Coding Intro
- Agentic Coding Handbook
Distillation¶
Evaluation¶
- Chatbot Arena: Benchmarking LLMs in the Wild
- DeepEval: Unit Testing for LLMs
- LLM Evaluation
- Auditing the Ask Astro LLM Q&A app
- giskard: Open-Source Evaluation & Testing for LLMs and ML models
- AI models collapse when trained on recursively generated data
- Beyond Traditional Testing: Addressing the Challenges of Non-Deterministic Software
- Introduction to Large Language Models
Explanation¶
- What is ChatGPT doing and why does it work
- Large language models, explained with a minimum of math and jargon
- Inside GPT: Understanding the text generation
- Understand how BERT constructs state-of-the-art embeddings
- From encoding to embeddings
- Large Language Models: Sentence-BERT
- Large Language Models: RoBERTa, a Robustly Optimized BERT Approach
- Generative AI exists because of the transformer: this is how it works
- All you need to know to Develop using Large Language Models
- A non-exhaustive but essential list of key papers that underpins text-to-video Deep Generative model like SORA
- Do large language models understand the world?
- Explaining generative language models to (almost) anyone
- The Rise of the LLM OS: From AIOS to MemGPT and beyond
- Will We Run Out of Data? Limits of LLM Scaling Based on Human-Generated Data
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- How LLMs Work, Explained Without Math
- Foundations of Large Language Models
- Transformers and Large Language Models cheatsheet for Stanford's CME 295
- Dummy's Guide to Modern LLM Sampling
- A cheat sheet for why using ChatGPT is not bad for the environment
- The Cultural Divide between Mathematics and AI
- 36 Alternatives to LLM Context
- Boring is good
- The security paradox of local LLMs
- Transformer Explainer: Interactive Learning of Text-Generative Models
- Text Tokens As Image Tokens
- Spec Driven Development
Foundation models¶
Framework¶
- LangChain: Building applications with LLMs through composability
- Cheshire-Cat: Production ready AI assistant framework
- OLMo: a State-of-the-Art, Truly Open LLM and Framework
- DSPy: the framework for programming - not prompting! - foundation models
- A programming framework for agentic AI
- Open Source Frameworks for Building Generative AI Applications
- LangChain for EDA: Build a CSV Sanity-Check Agent in Python
- Datapizza AI: a framework to build Gen AI agentic solutions
- Datapizza Buonanotte: Scientific Story Generator demonstrating DataPizza AI Agents
Implementation¶
- GPT in 60 Lines of NumPy
- privateGPT
- All You Need to Know to Build Your First LLM App
- How to Run LLMs Locally
- A Gentle Introduction to LLM APIs
- How to build a basic LLM GPT model from Scratch in Python
- Writing an LLM from scratch
- Teach your LLM about me
- nanochat by Andrej Karpathy: a full-stack implementation of an LLM like ChatGPT in a single, clean, minimal, hackable, dependency-lite codebase
Libraries¶
- StartChat Playground by Hugging Face
- DeclarAI: turning Python code into production-ready LLM tasks
- codellama
- Attention Sinks in LLMs for endless fluency
- OpenLLM Leaderboard
- LMQL: a programming language for large language models
- GPT-Engineer
- magentic: easily integrate Large Language Models into your Python code
- AlphaCodium: From Prompt Engineering to Flow Engineering
- Cohere For AI Launches Aya: an LLM Covering More Than 100 Languages
- Gemma: una nuova famiglia di modelli aperti
- Gemma 2 optimized for your local machine
- Unsloth: Finetune Llama 3.1, Mistral, Phi and Gemma
- Open WebUI: user-friendly WebUI for LLMs
- LangDrive: train LLMs on private data
- Trace: AutoDiff for AI Systems and LLM Agents
- LitGPT: 2high-performance LLMs with recipes to pretrain, finetune and deploy at scale
- guidance: a guidance language for controlling large language models
- litellm: Python SDK, proxy server to call LLM APIs using the OpenAI format
- guardrails: adding guardrails to large language models
- Burr: build applications that make decisions (chatbots, agents, simulations). Monitor, trace, persist, and execute on your own infrastructure
- el: a language model programming library
- JIT Implementation: A Python Library That Implements Your Code at Runtime
- ChainLit: Build Conversational AI in minutes
- DataChain: AI-data warehouse to enrich, transform and analyze unstructured data
- Simplemind: Python client for AI providers
- Docling: parse documents and export them to the desired format with ease and speed
- Posting: the modern API client that lives in your terminal
- TabPFN: Foundation Model for Tabular Data
- agx: AI Powered Analytics App
- torchexplorer: interactively inspect module inputs, outputs, parameters, and gradients
- token-explorer: a simple tool to explore different possible paths that an LLM might sample
- CoRT (Chain-of-Recursive-Thoughts): AI think harder when it argues with itself repeatedly
- agenticSeek: fully Local Manus AI
- blast: browser-LLM Auto-Scaling Technology
- Basic Memory: AI conversations that actually remember
- elroy: An AI assistant that remembers and sets goals
- langextract: A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization
- LLMs to Alloy
- Ollama Web search
- Llamafile: Distribute and run LLMs with a single file
- Agent Lightning: The absolute trainer to light up AI agents
- FastAPI MCP: Expose your FastAPI endpoints as Model Context Protocol tools
- TOON: Token-Oriented Object Notation
- Openskills: Skills assessment platform
MCP¶
- ask-human MCP
- fastmcp: The fast, Pythonic way to build MCP servers and clients
- Google A2A: Agent2Agent Protocol
- MCP As An Accidentally Universal Plugin
- Python MCP Server: Connect LLMs to Your Data
- Claude Skills are awesome, maybe a bigger deal than MCP
- What If You Dont Need MCP
- No OAuth Required MCP Client AWS IAM
Methods¶
- LLM sampling
- Open Source LLMs To Power A LLM Application
- NLP tasks via LLM
- Methods For Improving Your Large Language Model
Prompt engineering¶
- Pushing Prompt Engineering to the Limit
- Mastering Prompt Engineering
- The Prompt Engineering Playbook For Programmers
- Promptz.dev
RAG¶
- GraphRAG: a modular graph-based Retrieval-Augmented Generation (RAG) system
- llmware: unified framework for building enterprise RAG pipelines with small, specialized models
- talkd/dialog: RAG LLM Ops App for easy deployment and testing
- A RAG from scratch to query the scikit-learn documentation
- Production RAG: what I learned from processing 5M+ documents
Regulation¶
Vector databases¶
Visuals¶
- A Visual Guide to Mamba and State Space Models
- A Visual Guide to Quantization
- "Attention, Please!": A Visual Guide To The Attention Mechanism
- Official code repo for the O'Reilly Book "Hands-On Large Language Models"
- Understanding Transformers Using A Minimal Example