80% of a 110-node voice bot is repair logic: how to structure large flows?
RasonYang · reddit · 2026-10-27
A developer with production voice bots describes a structural wall: in a 110-node flow, 88 nodes (80%) were confirm/clarify/re-ask/fallback repair logic; only 15 were business logic. Canvas shows N edges but callers can trigger N² transitions, making them implicit via global handlers. Splitting into separate bots solves ownership, not complexity — handoffs add dead air and context loss. Sub-flows with typed inputs and a deterministic skeleton pushed the wall back but created the next one: ownerless global state like $confirmed. Asks the community about structure, repair logic placement, state scoping, and tooling (diffable DSLs, transcript-based regression tests).
More from coding & agent
- Alan AI launches to stop developers babysitting their coding agents — Scobleizer · 2026-10-27
- Compound Engineering ships ce-sweep: turns Slack/GitHub feedback into an executable dev plan — kieranklaassen · 2026-10-27
- Mintlify: doc traffic up 10x in a year, agent traffic up 135x, 60% of edits agent-written — handotdev · 2026-10-27
- Training Agents series wraps at GitHub SF: 4 talks on post-training open-model agents — SergioPaniego · 2026-10-27
- Grok Bot usage guide: event-based triggers and clear prompts beat routine wake-ups — XFreeze · 2026-10-27
- Dev builds Netlifyi, a Grok bot that digests Netlify changelogs daily at 9am — thisiskp_ · 2026-10-27