Not sure if this came through in the post, but with Rowboat we’re taking a strong stance: flowchart-style agent builders are useful today, but we believe they won’t scale with how LLMs are evolving.

As models get better at reasoning, you shouldn’t need to manually draw structured paths. It should feel more like onboarding a new teammate - you give high-level goals, and context, and they figure out the details. We don’t give flowcharts to teammates because it’s a lot of overhead to specify everything upfront. We think agentic systems are heading the same way. Flowcharts are helpful in some cases, but not how we’ll build long-lived assistants.

▲

kodefreeze 5 days ago | parent [-]

How do you build trust in a system like that? The flowchart style have the advantage that you can decide when you want a human to review/approve as well as ensuring actions that need to happen at certain conditions do happen.

	▲	segmenta 5 days ago \| parent [-]
		Yeah, flowchart style does have that advantage since you can add in approvals and conditions. The tradeoff is you end up limited to simpler logical flows that are easier to verify. Our take is that trust in agent systems has to be empirical. You start with manual testing and then layer on AI based simulations (we’re adding this in Rowboat soon) to test more scenarios at scale. Splitting work into multiple agents also makes it easier to isolate and test parts separately.