Ball: Complex sub-agent workflows will result in user hangovers
Thorsten Ball · Amp: The Emperor Has No Clothes · Sep 25, 2025 · at 35:22
Thorsten Ball, lead engineer on Sourcegraph's Amp coding agent, explains why brittle multi-agent architectures fail when faced with non-deterministic model outputs.
0:00 / 0:34exact quote · 34.1s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“A lot of the features what we see, you know, where people build like elaborate workflows, like I have my custom slash commands and they trigger custom Custom sub-agents and they in turn trigger custom MCP tool calls behind which again another model is doing inference again and taking the input and blah blah blah. I think a lot of this will and has resulted in hangovers where people realize, oh, like this looks like it's a deterministic workflow. It looks like it does the thing that I want it to do, but actually I can't use it if it only does it in 98% of the time.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Thorsten Ball
Ball: Amp core team has 8 people, ships 15 times daily without code reviews
“I think we're around eight people now on the AMP core team, and we still don't do formal code reviews. We still push to main. We still ship 15 times every day.”
Ball: AI coding tools must optimize harness and scaffolding per model
“And people are now also waking up to the fact that it's not just the model, it's the system prompt, it's the tools, it's the harness. The scaffolding around the model. So I can give you the choice to use Gemini 2.5 in, in AMP, but without the system prompting …”
Ball: Foundation models will become background implementation details in AI tools
“So I think we're going towards a future where the model will become an implementation detail to some sense, and we will end up on a different abstraction layer.”
Ball: Coding agents should outsource typing, not engineering thinking
“The failure mode of outsourcing the thinking, but not the typing, which I think it should be the opposite. You still have to know engineering. You still have to know how to program. You still have to know your application and its architecture, how it's deploye…”
Ball: Custom MCP tools fail if workflows diverge from frontier training
“If I give it this other custom-made MCP that we built internally, and our processes don't map to anything that OpenAI and Anthropic have seen or trained for, it won't be used, and you won't get good results.”
Ball: Developers are actively modifying codebases to suit AI agents
“What we're seeing now with agents is as soon as somebody has seen what it can do, they have such a multiplying effect or this brings so much value that people are willing to adopt the code base for this. Like the first time in how many decades where people are…”