Coding Agents Don't Scale Themselves. Neither Do Your Teams. — Patrick Debois, Tessl

By ai.engineer

Categories: AI, Tools

Summary

Coding agents won't differentiate your company—organizational readiness will. Patrick Debois argues teams must shift from fixing agent-generated code to improving the systems (context, harnesses, loops) that instruct agents, mirroring the continuous delivery adoption curve of 2009.

Key Takeaways

  1. Stop fixing code produced by agents; instead, improve the system that instructs them through better context engineering, harnesses, and loops. This mindset shift from tactical fixes to systemic improvement is the real competitive advantage.
  2. Developers resistant to agent work are your best assets for building better systems. Redirect their skepticism toward creating improved context and harnesses rather than dismissing them as blockers.
  3. The developer role evolves from coder to agent conductor/orchestrator, but pure prompt-writing creates friction. Reframe the work as 'tooling for agents' and 'context engineering' to retain technical engagement and craft.
  4. Coding agents are following the 2009 continuous delivery adoption pattern—organizations say 'it won't work here,' but they actually mean 'we're not ready yet.' Technology commoditization is inevitable; organizational readiness is the limiting factor.
  5. Apply engineering rigor to agent instructions the way you would for human engineers: require tests, documentation updates, and best practices in prompts. This prevents 'yolo prompting' and scales quality across autonomous systems.

Related topics

Transcript Excerpt

[music] >> Well, welcome. Um last day, I guess. That's what happens. Um I'm going to talk to you maybe not on the technical side, but more on the organizational side. So, if you're here for any technology, you can still leave if you want to. So, in 2009, um a lot of people were telling me the idea of continuous delivery was crazy. And I feel we're in kind of the same era or kind of the same thing right now with a dark factory. It will not work here. That's what I keep hearing over and over again. Um but what they're actually signaling to me, we're not ready yet. So, it's not the technology that can't make it work. It's not something they won't be able to do eventually, but they're just not set up for this. And there's been a lot of conference talks here about optimizing agents with loops a…

More from ai.engineer