Free Lesson
Five Mistakes Everyone Makes When Building AI Agents
45 min
Sep 17, 2026 7:00 PM
By continuing, you agree to Maven's Terms and Privacy Policy.
What you'll learn
Avoid the architecture mistakes that derail agents
Decide when you need a model call, a fixed workflow, or an agent loop, and why copying a coding agent can mislead you.
Catch the harness mistakes that add needless complexity
Spot unnecessary tools and orchestration, and see when model upgrades mean removing scaffolding instead of adding more.
Avoid the eval mistake that hides agent failures
Use real tasks, tool-call traces, and evals to catch failures and test whether changes to your harness actually help.
Why this topic matters
You add memory because coding agents have it, give the model control over steps your code could handle, and keep every workaround when you upgrade the model. Then you change the prompt, tools, and architecture together and wonder what helped. I'll unpack five mistakes in agent architecture, harness engineering, and evaluation, showing how to build less, inspect failures, and verify improvements.
You'll learn from
Hugo Bowne-Anderson
AI & data engineer, consultant, educator of 6+ million students (ex-Yale)
COACHED TEAMS AT
