GTM automation systems I run in production, AI agents I built for myself, and open-source tools anyone can use. Everything here exists and runs — nothing is a mock-up.
Seven repos of GTM automations plus Trigger.dev jobs, all pulled into one day-timeline: upcoming, ran, failed, never started. One-click re-run. A copy-context-for-debugging button. And an n8n watchdog that auto-dispatches any run GitHub Actions silently drops — the failure mode nobody notices until the data is stale. Built in one Claude Code session, live on Railway.
Related post: One of our “daily” GTM workflows was silently skipped 11 times in 30 days →A personal agent connected by webhooks. Every day it surfaces 5 random items from a 500–600-item GTM learning hub into Slack, receives captures from a Chrome extension into Notion, and translates my morning brief into a new voice via Sarvam.
Related post: The most underrated tool in my AI workflow right now is not Claude Code, Codex, or any agent — it is Webhooks →Captures each day’s Claude Code sessions, Granola meeting notes, and Slack threads into markdown. A weekly GitHub Action synthesizes what shipped and what’s pending to Confluence and Slack; a Hermes coaching note flags drift and repeating patterns.
Related post: Do you recall what you were working on during the week of May 1, 2026? →The production layer at Everstage: account scoring refresh, engagement intent reports, BDR performance tracking, outreach gap detection, and intel & reply extraction — running as daily crons.
Outbound replies land in an n8n and Trigger.dev flow where an LLM classifier labels intent and drafts a response. A human approves it in Slack before it goes out.
Slack AI agents that answer GTM questions for reps, with an eval suite so prompt and model changes are tested before they ship.
A pgvector retrieval layer over GTM content that grounds both rep questions and agent answers.
Want to talk through any of these?
Book a call