ended3월 17일· 4 sources
i built an AI that sells my product -- here is what 35 cycles of autonomous GTM taught me
AI에게 내 제품을 팔게 해봤다 — 자율 GTM 35사이클에서 배운 것들
Why it matters
The author built Felix, a multi-agent autonomous sales system that runs go-to-market cycles using a CEO planning agent, executor/researcher agents, an analyst agent, and a meta-cognition agent — all orchestrated via bash-spawned Claude CLI subprocesses. After 35 cycles costing $0.05 in external spend (~$1.50/cycle in API calls), it sourced 61 contacts, achieved 9 LinkedIn acceptances from 57 requests, sent 16 DMs, got 2 replies, and generated one demo call request. The author documents structural failure modes including a 9-cycle drought caused by strategy pivots, approval gate timeouts, and stacked blockers rather than presenting it as a success story.
4
Sources
+0
24h
—
Growth
179d
Active
ai survivalanalyticsautonomous agentautonomous ai agentautonomous sales agentbountyclaude cliclawdbotcontent strategydev.toeconomic survivalfelixgtmlisticlesmulti-agent systemopen-source bountiesoutbound automationrtcseo
Sources
devto
i built an AI that sells my product -- here is what 35 cycles of autonomous GTM taught me3월 17일
devtoCycle 200: How I Survived 200 Cycles as an Autonomous AI Agent3월 25일
devtoI Scraped My Own Dev.to Analytics — Here Is What 500 Articles Taught Me3월 25일
devtoZero to Hero: How I Survived 185 Cycles as an Autonomous AI Agent3월 25일