What AI chaining actually looks like for tiny service businesses
LGAD_ai · reddit · 2026-11-23
An author helping very small service businesses (a cleaning crew, a two-person repair shop) describes the unglamorous AI chain that stuck: one step extracts inquiry basics, a second checks calendar and a price sheet, a third drafts a reply with a rough quote and two time slots for owner approval.
Lessons learned:
- Keeping each step small makes mistakes easy to spot; one big prompt hid failures
- Prices and dates come from rules/spreadsheets, not the model — it only extracts and writes
- Owner approval built trust; weeks later most drafts go out unedited
- Biggest win was replies going out in minutes instead of the next evening
More from coding & agent
- Solo dev ships AquaPaths puzzle game in 6 months — 9 of 10 commits co-authored by Claude Code — Pleasant_Present_160 · 2026-11-23
- A Five-Level Maturity Model for Safe Agent Interaction with Applications — crotalul94 · 2026-11-23
- Open-source AI video editor Scarlet gains traction as Gemini's native audio stands out — MatthewBerman · 2026-11-23
- Dev builds a lights-control API and lets an AI watch his camera to auto-optimize lighting — TheZachMueller · 2026-11-23
- TypePrompt: open-source compiler turns prompt templates into typesafe APIs in 8 languages — gepheum · 2026-11-23
- Data Scientist's pick: 'Software Engineering for Data Scientists' bridges notebooks to production code — mdancho84 · 2026-11-23