Free Lesson
Forget Agents! Let's build a workflow instead
60 min
Mar 5, 2026 6:00 PM
Virtual (Zoom)
In this video
What you'll learn
Why fully autonomous agents often fail in production
Learn the common failure modes of autonomous agents, including drift, hidden errors, and evaluation blind spots
When to use agents vs structured workflows
Use a practical decision framework to determine the right level of autonomy, supervision, and control for your AI system
A lightweight evaluation harness you can implement now
Discover a simple evaluation and observability pattern you can add immediately to measure performance and catch failures
Why this topic matters
Agent this. Agent that. Aren't you tired of hearing about agents nonstop? Agents have become a little full of themselves.
Agents go off in left field to decide how THEY want do your task. But a lot of our LLM-assisted work requires predictability and determinism. In this talk we'll convince you to think outside the box to broader ways of doing agentic work.
You'll learn from

Hugo Bowne-Anderson
AI & data engineer, consultant, educator of 6+ million students (ex-Yale)

Doug Turnbull
Retrieval Consultant + Snuggie Connoisseur
Previously at
.png&w=1536&q=75)