Links I send people
Library
Pages I open myself and send to others: interactive explanations of how models work, playgrounds, roadmaps, guides on agents and evals, and the best material on tracking and estimation. Each has one line on what you will see or learn there.
LLM internals
See what happens inside a transformer, from tokens to the next-token probability.
- LLM Visualization (opens in a new tab)bbycroft.net
Fly through nano-GPT, GPT-2 and GPT-3 in 3D and follow one token through every layer and matrix multiply.
- Transformer Explainer (opens in a new tab)poloclub.github.io
Type a prompt into a live GPT-2 running in the browser and watch embeddings, attention heads and the temperature-adjusted output distribution change.
- Attention in transformers, step by step (3Blue1Brown) (opens in a new tab)3blue1brown.com
An animated lesson that shows queries, keys and values as geometry, so the attention formula stops being symbols.
- The Illustrated Transformer (opens in a new tab)jalammar.github.io
Diagram-by-diagram walk through the original encoder-decoder transformer, one tensor shape at a time.
- The Tokenizer Playground (opens in a new tab)huggingface.co
Paste text and compare how the tokenizers of GPT, Llama and other models split it, with token counts side by side.
- The Annotated Transformer (opens in a new tab)nlp.seas.harvard.edu
The 'Attention Is All You Need' paper interleaved line by line with runnable PyTorch code.
- How To Scale Your Model (opens in a new tab)jax-ml.github.io
Worked arithmetic for how transformers map onto accelerators: memory, FLOPs, and which parallelism to use when.
- On the Biology of a Large Language Model (opens in a new tab)transformer-circuits.pub
Attribution graphs that trace which internal features a production model uses for multi-step reasoning, planning rhymes and arithmetic.
Embeddings and retrieval
How text becomes vectors, how to look at those vectors, and how retrieval is built on them.
- Embedding Projector (opens in a new tab)projector.tensorflow.org
Load word2vec or MNIST embeddings and rotate them in PCA, t-SNE or UMAP to see which points end up as neighbours.
- Understanding UMAP (opens in a new tab)pair-code.github.io
Drag n_neighbors and min_dist on toy datasets and a 3D mammoth to see what UMAP keeps and what it distorts.
- Sentence Transformers documentation (opens in a new tab)sbert.net
The library docs for computing, fine-tuning and evaluating text embeddings and rerankers, with semantic search examples.
- MTEB Leaderboard (opens in a new tab)huggingface.co
Compare embedding models by task type, language, size and memory before picking one for retrieval.
- Contextual Retrieval (opens in a new tab)anthropic.com
Measured retrieval failure rates for plain chunking versus chunk context, BM25 hybrid search and reranking.
Agents and tools
Patterns, protocols and failure modes of LLM systems that call tools and act in loops.
- Building Effective Agents (opens in a new tab)anthropic.com
Workflows versus agents, five composable patterns, and when to skip frameworks and keep the loop simple.
- LLM Powered Autonomous Agents (opens in a new tab)lilianweng.github.io
A structured survey of planning, memory and tool use in LLM agents, with the papers behind each idea.
- Model Context Protocol specification (opens in a new tab)modelcontextprotocol.io
The current MCP spec: how a client and server negotiate tools, resources, prompts, auth and transport.
- Writing effective tools for agents (opens in a new tab)anthropic.com
How to design, prototype and evaluate tool definitions so a model picks and calls them correctly.
- Effective context engineering for AI agents (opens in a new tab)anthropic.com
What to put in a finite context window over a long run: compaction, note-taking, sub-agents and just-in-time retrieval.
- How we built our multi-agent research system (opens in a new tab)anthropic.com
A production write-up of an orchestrator with parallel sub-agents, including token cost, prompting lessons and how it was evaluated.
- 12-Factor Agents (opens in a new tab)github.com
Twelve engineering principles for LLM software that has to run reliably in production: own your prompts, own your control flow, and so on.
Evals and benchmarks
How to measure whether an LLM product works, and what public benchmarks show.
- Your AI Product Needs Evals (opens in a new tab)hamel.dev
A case study of building unit tests, human review and LLM-as-judge for a real product, and why that loop drives every improvement.
- AI Evals FAQ (opens in a new tab)hamel.dev
Direct answers to the common questions: error analysis first, binary judgments over scales, how many examples, when to trust a judge.
- Evaluating the Effectiveness of LLM-Evaluators (opens in a new tab)eugeneyan.com
A review of the research on LLM-as-judge: which setups agree with humans, known biases and how to measure a judge.
- Demystifying evals for AI agents (opens in a new tab)anthropic.com
How to grade multi-step agent runs: tasks, trials, graders and transcripts, and when to use code, model or human grading.
- Inspect (opens in a new tab)inspect.aisi.org.uk
The UK AI Security Institute's open-source framework for writing evals with datasets, solvers, scorers, tools and sandboxes.
- Berkeley Function Calling Leaderboard (opens in a new tab)gorilla.cs.berkeley.edu
See how models compare on tool calling across single, parallel, multi-turn and agentic categories.
ML intuition
Small models in the browser for building a feel for training, convolution and generation.
- A Neural Network Playground (opens in a new tab)playground.tensorflow.org
Train a tiny network in the browser and watch the decision boundary form as you change layers, features and regularisation.
- CNN Explainer (opens in a new tab)poloclub.github.io
Click any neuron of a small image classifier to see the kernel, the dot products and the activation map it produces.
- Diffusion Explainer (opens in a new tab)poloclub.github.io
Follow Stable Diffusion from prompt to image and see how guidance scale and seed change each denoising step.
- MLU-Explain (opens in a new tab)mlu-explain.github.io
Visual essays from Amazon on precision and recall, ROC curves, cross-validation, bias-variance and double descent.
- Why Momentum Really Works (opens in a new tab)distill.pub
Move the step size and momentum sliders and see gradient descent converge, oscillate or diverge on a quadratic.
- Understanding Deep Learning (opens in a new tab)udlbook.github.io
A free deep learning textbook by Simon Prince with 68 fill-in-the-code notebooks and a set of interactive figures.
Explorable explanations
Essays where the figures are live and the reader runs the experiment.
- Bartosz Ciechanowski (opens in a new tab)ciechanow.ski
Long essays on GPS, cameras, mechanical watches and airfoils where every figure is a 3D model you can drag.
- AI Explorables (Google PAIR) (opens in a new tab)pair.withgoogle.com
Short interactive pieces on memorisation versus generalisation, calibration, sparse autoencoders and fairness metrics.
- Seeing Theory (opens in a new tab)seeing-theory.brown.edu
Probability and statistics by simulation: roll dice, draw samples and watch the law of large numbers and Bayes' rule play out.
- Explained Visually (opens in a new tab)setosa.io
Interactive explanations of PCA, eigenvectors, Markov chains, conditional probability and image kernels.
- Distill (opens in a new tab)distill.pub
The archive of the machine learning journal where articles were interactive essays on feature visualization, attention and optimisation.
Computer vision
Vision and detection foundations, plus the tooling used in practice.
- Hugging Face Community Computer Vision Course (opens in a new tab)huggingface.co
A free course from CNN basics to vision transformers, detection, segmentation and multimodal models.
- CS231n: Deep Learning for Computer Vision (opens in a new tab)cs231n.github.io
Stanford's notes on image classification, backprop, CNNs and training practice, written to be read without the lectures.
- Computer Vision: Algorithms and Applications (opens in a new tab)szeliski.org
Szeliski's free second-edition textbook covering image formation, features, motion estimation, stereo and recognition.
- Ultralytics YOLO tracking mode (opens in a new tab)docs.ultralytics.com
How to run BoT-SORT or ByteTrack on top of a YOLO detector, with the tracker config options explained.
- supervision (opens in a new tab)github.com
A Python library of reusable pieces for detection pipelines: box and mask annotators, zone counting, ByteTrack and dataset conversion.
Tracking and estimation
Kalman filters, multi-target tracking, and how trackers are scored.
- Kalman Filter Explained Through Examples (opens in a new tab)kalmanfilter.net
Numerical examples, from weighing a gold bar to tracking a vehicle, that derive each Kalman filter equation before stating it.
- How a Kalman filter works, in pictures (opens in a new tab)bzarg.com
Gaussian blobs multiplied together show why predict and update give the Kalman equations.
- Kalman and Bayesian Filters in Python (opens in a new tab)github.com
A free book of notebooks that builds g-h, Kalman, EKF, UKF and particle filters from intuition, with simulations you can run.
- Stone Soup tutorials (opens in a new tab)stonesoup.readthedocs.io
Tutorials of an open tracking framework: Kalman and particle filters, data association, JPDA, GM-PHD and track metrics, with plots.
- TrackEval (opens in a new tab)github.com
The reference implementation of HOTA, CLEAR MOT and identity metrics used by the main tracking benchmarks.
- HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking (opens in a new tab)arxiv.org
Why MOTA over-weights detection and IDF1 over-weights association, and how HOTA splits the score into both.
- MOTChallenge MOT17 (archive) (opens in a new tab)motchallenge.net
The archived MOT17 leaderboard with 219 methods, per-sequence results and a CSV of every metric.
- ByteTrack (opens in a new tab)github.com
Code for the tracker that also associates low-confidence detections, which recovers occluded objects that score-thresholding drops.
Roadmaps and courses
Structured paths, from a first map of the field to building a model from scratch.
- AI Engineer Roadmap (opens in a new tab)roadmap.sh
A clickable map of what an AI engineer is expected to know, from model APIs and embeddings to RAG and agents.
- AI Agents Roadmap (opens in a new tab)roadmap.sh
A clickable map of agent topics: the agent loop, tools, memory, protocols, evaluation and security.
- Neural Networks: Zero to Hero (opens in a new tab)karpathy.ai
Karpathy builds backprop, a character model, a tokenizer and GPT from an empty file, in video with code.
- Build a Large Language Model (From Scratch) (opens in a new tab)github.com
Notebooks that implement a GPT-style model in PyTorch, then pretrain, fine-tune for classification and instruction-tune it.
- Stanford CS336: Language Modeling from Scratch (opens in a new tab)cs336.stanford.edu
Lectures and assignments that build tokenizer, transformer, training, scaling laws, data and alignment yourself.
Writing worth following
People who publish careful, first-hand notes on building with ML and LLMs.
- Simon Willison's Weblog (opens in a new tab)simonwillison.net
Daily hands-on notes on new models, prompt injection, tools and what he actually built with them.
- Lil'Log (opens in a new tab)lilianweng.github.io
Long literature reviews on agents, hallucination, diffusion, RL and prompting, each with the full reference list.
- Chip Huyen (opens in a new tab)huyenchip.com
Essays on ML systems in production, AI engineering and building LLM applications, from the author of 'AI Engineering'.
- Eugene Yan (opens in a new tab)eugeneyan.com
Applied notes on recommender systems, evals, LLM patterns and ML team practice from industry work.
- What We Learned From A Year of Building with LLMs (opens in a new tab)applied-llms.org
Tactical, operational and strategic lessons from six practitioners who shipped LLM products.
Every link was opened on 2026-09-26.