Traces and incident reviews: See it to fix it

In this class

What this class covers

1. 01Identify inputs, retrieved data, and tool calls to capture in execution traces
2. 02Distinguish issues involving data, models, tools, permissions, and workflows
3. 03Reproduce incidents in a test environment and repair the appropriate layer
4. 04Use retest results to choose relaunch, downgrade, rollback, or human handoff

Instructor

Learn with Sean Liu

Sean Liu

Sean Liu

CEO, Narsil AI

• Founder and CEO of Narsil AI; previously founded Botsnova
• Long focused on NLP, conversational AI, RAG, LLMOps, and AI agents
• Hands-on experience connecting AI to companies' existing systems, data, and workflows
• Helped manufacturers turn master craftsmen's know-how, quality judgment, and shop-floor data into executable standards

Evolve · 7 classes

7 courses in Evolve

One price covers all 7 classes. Start with the one you want.

1. 01 Loop engineering: Write acceptance criteria and let AI iterate to passHai Tseng · About 18 min →
2. 02 Turn taste into evals: External signals and machine-readable criteriaSean Liu · About 30 min →
3. 03 Make your team's AI better with use: Shared knowledge, SOPs and quality gatesCarmen Hsueh · About 20 min →
4. 04 AI governance boundaries: Permissions, cost and human handoffWisely Chen · About 35 min →
5. 05 Traces and incident reviews: See it to fix itSean Liu · About 30 min This course
6. 06 Self-evolving AI systems: Build your own upgrade systemHai Tseng · About 30 min →
7. 07 Five decisions for going all-in on AI: Alignment, process, business model, pricing and budgetSting Tao · About 60 min →