Twelve DeepLearning.AI agent courses taught by Anthropic, Google, LangChain and Andrew Ng, in the order to take them — every link, hours, prerequisites, a 4-week plan, and the free-vs-Pro catch table

The companion to the reel. DeepLearning.AI's twelve agent courses are free to watch during its platform beta; this map puts them in an order that builds: TIER 1 FOUNDATION — Andrew Ng's Agentic AI (about 10 hours, the four agentic patterns). TIER 2 TOOLS — MCP with Anthropic, Agent Skills with Anthropic, Long-Term Agentic Memory with LangGraph. TIER 3 BUILD — six courses, pick two: browser agents (AGI Inc), data agents (Snowflake), adaptive agents (Oracle), generative UI (CopilotKit), image and video agents (Google), coding workflows from cloud to local (JetBrains). TIER 4 SHIP — A2A, the Agent2Agent protocol (Google Cloud and IBM Research), and Evaluating AI Agents (Arize). PAGE 1 — the map, all twelve links with hours and levels. PAGES 2 to 5 — one page per tier: why each course is where it is, what is inside, what it needs. PAGE 5 also carries the 4-week plan. PAGE 6 — the catch table: watching is free, certificates and graded assignments need the Pro membership, and every course recommends Python. Facts verified September 13, 2026. Independent guide from Hyperautomation Labs — not affiliated with DeepLearning.AI or its partners.

By submitting, you'll be subscribed to the Hyperautomation AI Report newsletter — three practical AI emails a week. Unsubscribe anytime.

Free. No spam. Unsubscribe anytime.