Google releases free 2-hour course on Agent Harness Engineering, from prompt to self-improving agents
Roger_M_Taylor · x · 2026-09-19
Google has released a free 2-hour course on Agent Harness Engineering, walking from a single prompt to a self-improving harness.
The curriculum covers:
- Building your first AI agent (10:16)
- Mastering prompt engineering (41:05)
- Turning agents into graphs (54:45)
- Running loops inside an agent harness (1:20:10)
- Building a harness that improves itself (1:43:33)
The accompanying article lays out a 6-layer harness framework: a better prompt improves one answer, a better harness improves every run — addressing agents that forget constraints, pick wrong tools, skip verification, or loop endlessly.
More from coding & agent
- Hindsight ships 0.8B model for agent memory decisions: 2.5x cheaper, 1.9x faster — victor_explore · 2026-11-04
- Human Oversight Is 70-75% of AI Agent Costs in Banking, McKinsey Finds — mikeflache · 2026-11-04
- Over 80% of Major Financial Institutions Now Run Agentic AI in Production — mikeflache · 2026-11-04
- Mnemos MCP nears launch after shared memory layer fix — RileyRalmuto · 2026-11-04
- Moving mobile dev to remote machines makes it multiplayer across product, design, engineering — rudrank · 2026-11-04
- Kuaishou's EvoRec: self-evolving agentic recommender lifts ad revenue 4.6% — _reachsumit · 2026-11-04