AI Agents — resources
roadmap.sh: https://roadmap.sh/ai-agents
Books
- AI Engineering (Chip Huyen) — practical, vendor-neutral coverage of building applications on top of foundation models: evaluation, RAG, agents, and deployment.
- Building LLMs for Production (Bouchard & Peters) — hands-on guide to RAG, fine-tuning, and agentic patterns with real framework code.
- Anthropic — “Building Effective Agents” — the canonical engineering essay on agent vs. workflow patterns (prompt chaining, routing, orchestrator-workers, evaluator-optimizer). Read this first: https://www.anthropic.com/research/building-effective-agents
- ReAct: Synergizing Reasoning and Acting in Language Models (Yao et al., 2022) — the paper behind the reason+act loop most agents use: https://arxiv.org/abs/2210.03629
Courses / practice
- DeepLearning.AI — short courses on agents — free hands-on courses (Functions/Tools/Agents with LangChain, AI Agents in LangGraph, Multi-AI-Agent Systems with crewAI, MCP).
- LangChain & LangGraph docs — reference + tutorials for building stateful, multi-step agent graphs; the most-used Python/JS agent framework.
- Model Context Protocol (MCP) docs — the open standard for connecting agents to tools and data sources; includes spec, SDKs, and server quickstarts.