Awesome AI

End-to-end AI product building (Chip Huyen).

The canonical AI theory text (Russell, Norvig).

Transformers in raw PyTorch, layer by layer (Sebastian Raschka).

Mathematical foundations of neural networks (Goodfellow, Bengio, Courville).

Bishop's 2024 update; probability-grounded modern DL (Bishop & Bishop).

Scalable, maintainable ML pipelines (Chip Huyen).

GANs, VAEs, diffusion models (David Foster).

Visual + practical guide to LLM applications (Jay Alammar, Maarten Grootendorst).

Production LLMOps: fine-tuning, quantization, serving (Labonne, Iusztin).

RL foundations (Sutton, Barto).

The NLP reference, kept current through the deep learning era (Jurafsky, Martin).

Concise, math-grounded path from n-grams to transformers (Andriy Burkov).

Math + intuition + Python notebooks (Simon Prince).

Git-integrated pair programming with surgical edits and undo.

Sourcegraph's commercial agentic coding tool with strong product UX.

Anthropic's CLI agent; multi-file codebase refactoring with long context.

Open-source agentic IDE extension with strong multi-provider support.

OpenAI's Rust-based local terminal agent; lightweight and fast.

Open-source IDE + CLI assistant with source-controlled rules.

Cursor's terminal-native agent with sandboxed permissions.

Benchmark-leading multi-model harness with BYOK local execution.

Google's official open-source terminal agent; long-context repo exploration.

Block's extensible, MCP-driven local agent.

Multi-mode agent with a unified gateway to 500+ models.

Mistral's agentic coding CLI, powered by Devstral.

Private, local-first agent for Ollama and LM Studio.

Provider-agnostic terminal harness with a strong TUI.

Open-source autonomous SWE platform; browser + shell + editor loop.

Highly customizable terminal harness; minimal base prompt, extension-driven.

Alibaba's terminal coding agent, optimized for Qwen models.

OpenAI's framework for writing evals.

Microsoft's framework for multi-agent conversation and collaboration.

Agent orchestration with structured tasks and human-in-the-loop controls.

Great library for ingesting any kind of document for RAG ⭐

Google's Agent Development Kit (Python, Java). Great local development experience + A2A + MCP.

Open-source search/RAG framework with modular pipelines.

Build multi-agent workflows with stateful graphs on top of LangChain.

Data framework for ingesting, indexing, and querying private data with LLMs.

Extremely minimalist AI agent framework in just 100 lines of code. Fantastic way to learn.

Typed, structured LLM orchestration framework built on Pydantic models for safe, predictable outputs.

— ⭐ Patterns, pitfalls, and tradeoffs for designing AI agents.

Companion guide from Google.

Practical guide to building AI agents from Google.

A goldmine of useful links for AI engineers.

Practical guide on building agents.

Example code, recipes, and best practices for working with OpenAI APIs.

LLM-powered IDE for multi-file edits and codebase-aware chat.

In-IDE code completion, chat, and refactors.

Transformer architecture.

Safer model alignment.

GPT-3 capabilities.

Model/data/compute scaling.

High-quality text-to-speech and voice cloning.

AI music from text prompts.

High-quality video with synchronized audio.

Cinematic, realistic video generation.

Video editing + generation with granular creative control.

Best for general reasoning, tool use, and the broadest ecosystem.

Best for long-context analysis, coding, and structured thinking.

Best for enterprise LLMs with strong retrieval-augmented generation APIs.

Best for cost-efficient reasoning with open weights.

Best for multimodal tasks and Google ecosystem integration.

Frontier-tier Chinese model with open weights.

Best for real-time information via X and very long context.

Best for long-context instruction following.

Best open-weight family for self-hosting and fine-tuning.

Best for lightweight, high-performance open-weight models.

Best for multilingual and Chinese-first applications.

Speed, price, and quality benchmarks across providers.

Human-preference Elo rankings for text, image, and video.

Unified API + live pricing across ~300 models.

Integrated into Creative Cloud; commercial-safe.

High-res, prompt-editable, open-weight images.

OpenAI's integrated image generation with near-perfect text rendering.

Precise, legible text in generated images.

Artistic and photorealistic images.