AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Discover how to scale AI agents that autonomously handle thousands of research and automation tasks weekly through principled software engineering approaches in this 28-minute conference talk. Learn from Linus Lee's experience building Thrive Capital's in-house research engine, Puck, which executes thousands of tasks—from surfacing current events to drafting reports—through a robust orchestration library called Polymer. Explore how a team of fewer than 10 engineers sustains this scale by applying four core engineering principles: composability, observability, statelessness, and changeability. Examine the system's architecture through design patterns and practical implementation examples that make complex agentic systems maintainable, testable, and scalable. Understand how classical engineering discipline can co-exist with frontier LLM innovation, including design patterns for building robust, composable LLM systems at scale, applying principles like composition, adapters, and stateless effects in production, implementing observability and traceability in multi-agent orchestration, evolving systems seamlessly as new model capabilities emerge, and why classic software engineering values matter more than ever in the AI era.
