Skip to content

Machine-Learning-Tokyo/agent-resources

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

Learn how to build, deploy, evaluate agents (free resources)

A curated set of free, high-quality resources, ranked by overall quality and impact. Coverage spans the full lifecycle: learn → build → deploy → evaluate.

  1. Anthropic — Building Effective Agents (+ cookbook reference implementations) The canonical essay on the workflow-vs-agent distinction and composable patterns (prompt chaining, routing, parallelization, orchestrator-worker, evaluator-optimizer). Core principle: start simple, add complexity only when it pays off. Runnable code for every pattern.

  2. DeepLearning.AI — Agentic AI (Andrew Ng) (~6 hrs, free, certificate) The four design patterns (reflection, tool use, planning, multi-agent), taught with evaluation-driven development and systematic error analysis woven throughout. The best single structured course for going from concept to production discipline.

  3. OpenAI — A Practical Guide to Building Agents (free PDF) Written for product and engineering teams: when to build an agent at all, design foundations, orchestration patterns, tool design, and guardrails. Concept-level guidance is model-agnostic even though examples use OpenAI. Strong on the PM-facing scoping question.

  4. Hugging Face — AI Agents Course (free, certified) Community-driven, theory-to-practice. Build with smolagents, LlamaIndex, and LangGraph; experiment in Hugging Face Spaces; finish with a benchmark assignment and a public leaderboard. The most hands-on of the structured courses.

  5. DeepLearning.AI — Evaluating AI Agents (beta — free for now) The eval-specific entry, and the one most lists skip. Add observability via traces, then choose the right evaluator — code-based, LLM-as-a-Judge, or human annotation — for each component, evaluating both component-wise and end-to-end. (Built with Arize AI.)

  6. UC Berkeley CS294-196 — Agentic AI MOOC (Fall 2025) Free lecture videos and materials covering the academic and architectural foundations of agentic AI. Heavier and more research-oriented than the courses above — best for depth on the why behind the patterns.

  7. DeepLearning.AI — AI Agents in LangGraph (beta — free for now) Build an agent from scratch in Python, then rebuild it in LangGraph: agentic search, persistence and state across threads, and human-in-the-loop. The best free intro to controllable, stateful orchestration.

  8. Anthropic — Building Effective AI Agents: Architecture Patterns & Implementation Frameworks (free eBook) The deploy/production companion to #1. How to choose between single-agent, multi-agent, and workflow architectures, with real-world examples from companies running these at scale (Coinbase, Intercom, Thomson Reuters).

  9. Hugging Face — MCP Course (free) The Model Context Protocol from fundamentals to advanced, in theory, design, and practice. Narrower, but valuable: the tool/connector layer is where a lot of real agent-reliability problems actually live.

  10. Kaggle — 5-Day AI Agents Intensive (Vibe Coding) with Google A practical, hands-on course-and-competition focused on vibe-coding workflows, where natural language becomes the primary interface for building production-ready systems. Practical and fun; lighter on rigor than the rest.

About

Learn how to build, deploy, evaluate agents (free resources)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors