Continuous Production Performance Optimization via Agents
AI Engineer · youtube · 2026-07-19
This talk shares how Hud integrates runtime context into coding agents for continuous production performance optimization.
The core problem is that performance degradation in mature codebases often accumulates silently; teams know there might be issues but find it hard to pause feature development to investigate or estimate the time needed. Hud's approach involves having agents analyze real production environment data to automatically identify high-ROI optimization points, scoring them by complexity and impact to help tech leads prioritize the most worthwhile fixes.
In their case study, the system discovered and fixed N+1 queries and missing database indexes in its first week, delivering measurable P90 latency improvements post-launch. Tech leads can now receive actionable reports directly before sprint planning, shifting from "just knowing there's a problem" to "seeing exactly how to fix it."
More from coding & agent
- OpenAI ships GPT-6.1 Terra; Brockman says its computer use now beats his own speed — gdb · 2026-10-27
- Kolab Benchmark Uses Pro Screen Recordings to Show AI Agents Overestimate Collaboration Quality — minnesotanlp · 2026-10-27
- Dev argues Claude burns far more tokens than GPT models for the same coding work — haider1 · 2026-10-27
- Stop building agents that can't change models: stable contracts, external state, real evals — rseroter · 2026-10-27
- Cheap models can cost more in agent harnesses than frontier ones, researcher warns — vasuman · 2026-10-27
- Kanzo: A 5k-Line Rust Kanban Binary With MCP for Claude Code and Codex — allmudi · 2026-10-27