Richard Lander shows how SKILL.md files packed inside a NuGet package reach the NuGet cache on dotnet restore, giving agents guidance that matches the exact package version

180-page survey of mechanistic interpretability covering superposition, sparse autoencoders, circuit discovery, and steering, with a companion course and Jupyter notebooks

TypeScript meta-framework of spatiotemporal composability where plugins register services, typed events and revertible effects on one shared context, with reload and teardown undoing them.

LLM pipeline that maps topic hierarchies in AI conversation logs under differential privacy, without relying on embeddings.

MIT Press open-access textbook on evolving neural networks and agent designs, with hands-on exercises, review questions and an online supplement

Free book on serving generative AI models in production, readable online, covering GPU hardware, inference engines, techniques and operations

Eleven free machine learning and AI textbooks covering classic learning theory, deep learning, reinforcement learning, multi-agent systems, fairness and probabilistic methods

• Foundations of Machine Learning
• Understanding Deep Learning
• Introduction to Machine Learning Systems, Vol. 1
• Introduction to Machine Learning Systems, Vol. 2
• Algorithms for Machine Learning
• Deep Learning
• Reinforcement Learning (CMU 10-703)
• Distributional Reinforcement Learning
• Multi-Agent Reinforcement Learning
• Agents in the Long Game of AI
• Fairness and Machine Learning
• Probabilistic Machine Learning, Book 1
• Probabilistic Machine Learning, Book 2

Guided journey into abstract mathematics and proof writing, from logic and sets to induction, with exercises and a theorem appendix

Graduate-level graph theory textbook on trees, colorings, matching and network flows, with about 100 exercises and no solutions

Open-source ESP32 quadcopter project with Arduino firmware, Gazebo simulation, and a Python library for scripted flights.

End-to-end project that trains a ~9M parameter fish-themed language model on synthetic data in about five minutes.

Curated directory of open source alternatives to proprietary software.

Trainable key-value memory layers that add model capacity without increasing FLOPs.

Research code for large concept models that predict in a sentence representation space instead of tokens.

Reference code for the Byte Latent Transformer, a tokenizer-free byte-level language model architecture.

A 361-page reference covering 151 trading strategies with 550+ mathematical formulas, code for backtesting, and strategies using neural networks, Bayesian methods, and k-nearest neighbors

Anthropic's announcement of Claude Fable 5.1 and Mythos 5.1, with major gains in coding, agentic tasks, and research.

Research paper introducing reinforcement learning with metacognitive feedback (RLMF) to train LLMs to express uncertainty faithfully, aligning expressed with intrinsic uncertainty.

Google research paper proposing SKILL.state, which replaces full dialog and reasoning history with a compact structured execution state, cutting prompt tokens about 16x on long tasks.

Free 737-page mathematical introduction to deep learning covering linear algebra, optimization, backpropagation, CNNs, sequence models, and approximation theory.

Tencent WeChat's paper introducing WeMM-Embedding, a family of multimodal embedding models unifying text, images, video, and documents in one semantic space.

Chapter-by-chapter notes for Alex Xu's System Design Interview - An Insider's Guide, covering Volumes 1 and 2.

Notebook from Anthropic's Applied AI team walking through techniques to cut Claude API spend without sacrificing quality.

Preprint formalizing revertible effects and reactive coeffects into a programming paradigm for temporal and spatial composability, implemented in the Cordis meta-framework.

Explains Cursor's open-source MoE training megakernel for NVL72s that fuses communication and computation into a single deterministic kernel.

Defines an open, vendor-neutral format for packaging Agent Skills and MCP servers into distributable plugins for AI agents.

Guide to web development using only vanilla techniques — no tools, no frameworks, just web standards.

Curated collection of free legal documents and templates for companies, including site policies, agreements, and contracts.

Argues that enterprise-scale self-evolving LLM agents are held back by inadequate agentic online RL systems and proposes three architectural pillars for next-generation infrastructure.

Foundation model for tabular data classification and regression using in-context learning that eliminates manual feature engineering.

Curated collection of CS fundamentals for placement preparation covering DSA, computer networks, DBMS & SQL, OOPs, operating systems, system design, and software engineering.

Free curated technical interview preparation guide covering coding algorithms, system design, behavioral questions, and resume tips

Curated awesome list of interview question collections across programming languages and technologies

Curated collection of free resources for AI/ML security, prompt injection, and AI pentesting, including guides, CTFs, and academic papers.

Ai2 compares Olmo 3 (decoder-only transformer) and Olmo Hybrid (RNN/transformer hybrid at 3:1 ratio), finding hybrids better at predicting meaning-bearing tokens while transformers excel at verbatim copying.

Detailed technical article by Sber describing a managed Deep Research pipeline built with GigaChat Ultra 3.1 using Planner, Extractor, Validator, and Answer Builder roles for controlled multi-step research

Technical deep-dive into MiMo Code, Xiaomi's open-source terminal-based coding agent designed for long-horizon automated programming tasks with infinite logical sessions and four-level memory architecture

Deep dive into vLLM's architecture covering paged attention, continuous batching, prefix caching, speculative decoding, chunked prefill, and multi-GPU/multi-node distributed serving.

Practical guide to building production DSPy applications covering async RAG pipelines, GEPA prompt optimization, FastAPI serving, and LangSmith monitoring.

LeetCode for PyTorch — practice implementing operators and architectures from scratch with instant auto-grading.

A hands-on course to build an LLM serving system from scratch using Python and MLX, covering inference, optimization, and batching.

Notebooks, data, and supplementary material for the third edition of Think Stats.

Technical book chronicling the journey of scaling a startup from a single server to global infrastructure.

Explain complex systems using visuals and simple terms to help prepare for system design interviews.

Thoughtworks methodology treating prompts as first-class delivery artifacts with the REASONS Canvas to make LLM-assisted changes governable, reviewable, and reusable.

Completed research on semantic retrieval augmented generation using novel semantic similarity graph traversal algorithms.

Comprehensive guide to SHAP (SHapley Additive exPlanations) covering waterfall plots, force plots, summary plots, and model interpretability for tree-based and deep learning models.

Comprehensive collection of advanced Retrieval-Augmented Generation techniques with detailed notebook tutorials.

Open-source textbook and curriculum for machine learning systems engineering.

Implement a ChatGPT-like large language model in PyTorch from scratch, step by step.

Curated list of datasets and tools for post-training large language models focusing on accuracy, diversity, and complexity.

Course on making agentic coding tools reliable through environment design, state management, and control systems.

A nano Claude Code-like agent harness built from 0 to 1, teaching harness engineering through 12 progressive sessions.

Hands-on learning platform for Linux, Docker, Kubernetes, and networking with interactive tutorials, coding challenges, and sandboxed playgrounds.

Tutorial on fine-tuning LFM2.5-1.2B-Instruct with GRPO and Unsloth for structured data extraction from OCR text.

Wiki related files for Electronics in PTHS Wiki (schematics, drawings, and images).

Free companion website for the Deep Learning with Python book covering neural networks, TensorFlow, PyTorch, JAX, Keras, and real-world AI workflows.

Comprehensive guide to prompt engineering techniques for Claude's latest models, covering clarity, examples, XML structuring, thinking, and agentic systems.

Collection of notebooks and code snippets showcasing ways to build with Claude for developers.

Collection of step-by-step guides to rebuild popular technologies from scratch.

Curated list of awesome articles, videos, and other resources to learn and practice software architecture, patterns, and principles.

Curated list of artificial intelligence tools, models, and platforms across the evolving ecosystem.

Workshop for IT/Ops teams to learn how to manage AI-enabled applications on Microsoft Azure.

In-depth tutorials on LLMs, RAGs and real-world AI agent applications.

A curated collection of over 500 AI agent use cases across various industries with links to open-source projects for implementation.