5-Day AI Agents Intensive Course with Google
Google's intensive, now a free self-paced Learn Guide: daily codelabs build agents with ADK and Gemini, covering tools, sessions, memory, observability, evaluation, A2A, and deployment. We could not confirm a certificate for the self-paced version.
Go to the course on Kaggle- Provider
- Google and Kaggle
- Price
- Free
- Certificate
- Certificate not statedThe Learn Guide page does not mention a certificate or badge for the self-paced version.
- Level
- Intermediate
- Time
- 5 days of material
- Format
- Self-paced Kaggle Learn Guide: whitepapers, summary podcasts, and Kaggle notebook codelabs (originally a live 5-day course in November 2025)
- Checked
- 2026-11-06
Needs a Kaggle account (phone-verified to run notebooks) and a Gemini API key from Google AI Studio.
01
What you will learn.
- 01Agent Development Kit (ADK)
- 02Gemini
- 03Agent tools
- 04MCP
- 05Sessions and memory
- 06Observability
- 07Evaluation
- 08Agent2Agent (A2A)
- 09Deployment
02
Before you enroll.
From the course page, checked 2026-11-06.
- Who it is for
- Developers who want to learn the foundations and practical use of AI agents.
- You need
- A phone-verified Kaggle account
- A Google AI Studio account that can create a Gemini API key
- What you build
- A first agent with ADK and Gemini that uses Google Search
- A multi-agent system built with ADK
- Agents with session memory and long-term memory
- Agents that work together over the A2A Protocol
- Good to know
- The capstone was a separate November 2025 live-course challenge; self-study access does not confirm current submission or badge eligibility.
- The livestreams ran in November 2025; only recordings are left
- The optional deployment codelab runs on Google Cloud Vertex AI Agent Engine
03
Similar courses.
Agentic AI
Andrew Ng teaches the core agentic design patterns (reflection, tool use, planning, multi-agent) in plain Python before any framework. The best conceptual course here; the certificate requires Pro.
Free + paidPaid certificateIntermediate
Evaluating and Improving Agentic Applications
The eval course most agent tutorials skip: traces, test sets, code evaluators and LLM judges, judge reliability, and turning production failures into regression tests. Take it once you have an agent worth measuring.
PaidPaid certificateIntermediate
AI Agents Course
The best free starting point: agent theory first, then the same ideas in smolagents, LangGraph, and LlamaIndex. Two free certificates: a fundamentals one after Unit 1 and a completion one after the use-case assignment and final challenge.
FreeFree certificateBeginner