Summary
Most scale-ups hit the same wall. In startup mode, debt grows slower than headcount, so debt per developer stays low and teams keep moving. As the company scales, debt per developer climbs and output drops; teams cut standards to chase output, buy a few months, then hit the wall of outages and angry customers. AI runs this same pattern at record speed.
This roundtable is about getting ahead of it: the measures, practices, and incentives that make debt visible before it compounds, why rewarding the absence of drama beats rewarding raw output, and how to apply AI to surface and pay down debt rather than only generate more. Come with what's worked and where it's burned you.
Host
Peter Edworthy
Director of Engineering - Web Platform @ BitSight
Peter Edworthy leads Platform & Core Engineering at BitSight and has driven agentic coding adoption across 200+ engineers, working hands-on with LLM code generation at scale. He writes on engineering leadership in the AI era, arguing that AI should lower the cost of change rather than bury it as debt.
Sign up to receive news and updates on ELC Annual 2026
