Episodes (Page 3)
✨
Plain grep is insufficient for searching vast, messy company data.
✨
Engineers are shifting from coding to governing AI agents.
Davis Palmie
✨
Self-improving agents can share learned skills across an organization.
Rafal Wilinski
✨
AI attackers are faster than traditional pen tests.
✨
AI agents currently require manual message routing between them.
Vlad Luzin
✨
Agents should run in isolated cloud environments, not local machines.
✨
AI companies face unexpected infrastructure and pricing challenges.
Dor Sasson
✨
AI excels at writing code but struggles with production execution.
Varun Krovvidi
✨
Speculative decoding uses a small model to speed up large model inference.
Sheilah Kirui
✨
AI agents significantly increase traffic, challenging database scalability.
✨
Measuring agent quality is complex, extending beyond spreadsheets.
Hossein Niazmandi
✨
AI can automate agent improvement loops, reducing manual effort.
Marc Klingen
✨
Harness engineering builds software factories run by agents.
Dru Knox
✨
Dashboards are dead; users interact via AI, not static views.
Sarah Simionescu
✨
Fine-tuning isn't the fix for retrieval issues; architecture is key.
Anant Srivastava
✨
Dan Adler of Sourcegraph discusses AI coding agents impacting large codebases.
✨
Sandhya Subramani from AWS demos Strands Agents SDK for runtime tool creation.
✨
Nixon Dinh from PayPal explains the shift from keyword to AI agent search.
Nixon Dinh
✨
Melanie Warrick from Temporal on building human-in-the-loop agents.
Melanie Warrick
✨
Matt Lawler shares AssemblyAI's AI support agent, Joey, resolving 80% of tickets.