In Part 5 of the SambaNova coding agent series, Kwasi Ankomah puts traces to work. He analyzes the agent’s failing runs, diagnoses the recurring failure modes, and then updates the agent’s instructions, memory, and tools automatically — using off-the-shelf libraries rather than a custom pipeline.
WebinarUpcoming
Build self-improving AI agents: Turn failing traces into higher pass@k
This is the heart of the series — and the step most teams skip. Kwasi Ankomah shows how to read an agent's failing runs, find the patterns behind them, and automatically improve the agent without any fine-tuning.
Explore more events and resources
Browse our library of tutorials, webinars, and live sessions to keep building your AI skills.