AI GTM Enrichment Pipeline

Three versions.
One problem.
Not finished.

Fleet Engine is an AI system that finds companies that are likely to need your service, researches exactly what's broken inside each one, identifies a real person to talk to, and writes a first message tailored to their specific situation — not a template, not a blast, one message per company grounded in evidence. It reads public job postings as the signal: a company hiring for RevOps, or bringing in a financial controller, or scaling a growth team is showing you exactly what problem they're paying to solve right now. Fleet Engine reads that signal, enriches it, and turns it into a warm outreach in minutes. This is the engineering record of building it across three versions until the infrastructure stopped failing.

Select a version to enter
The Fleet
🐋 Ahab
The Hunter
Bypasses noisy job board aggregators and goes straight to the source — company careers pages, Lever, Greenhouse, Ashby. Runs 5+ targeted search queries per campaign, cross-references hiring signals with funding events, and surfaces companies actively in motion. Every prior discovery is excluded automatically, so each run is a net-new list.
🔬 Nemo
The Analyst
Dives deep into each company for the cost of a Vertex AI API call. Verifies the real company domain (not the job board URL), identifies a named decision-maker with a real LinkedIn profile, and classifies the operational friction behind why they're hiring. Any lead where the most recent funding event is older than 18 months gets cut — stale signals make bad outreach.
🔱 Neptune
The Copywriter
Writes the first message. Built on the methodology of Eugene Schwartz — a 1960s direct response copywriter whose book Breakthrough Advertising still sells for $500 used. Schwartz's rule: you cannot create desire in the market, only channel what already exists. Neptune reflects what the prospect is already doing, names the specific tool or process failing them, and offers the exact operational outcome they already want. Three sentences. No fluff.
V1 · Archive
Fleet Engine: Fathom

Three agents over n8n, Postgres, and Docker on GCP. 1,500 stress-test cycles across six campaigns. Broke 18 times. Every failure documented forensically before the rebuild.

3 agents 15 funded leads 18 failure modes 1,500 cycles
V2 · Live
Fleet Engine: Cortex

Claude Code as Admiral. ADK on Agent Engine. GCS file handoff. Three Vertex AI RAG corpora compounding across runs. Clay delivery. Same agents, new substrate.

Python ADK GCS handoff 3 RAG corpora Clay → HubSpot 0 self-hosted
V3 · Locked
Self-Improving Pipeline

Reply outcomes flow back into the RAG store. A fourth agent uses that signal to improve outreach before delivery. The pipeline learns what converts.

Polish agent Outcome feedback loop RAG compounds Reply signal
🔒
Locked
Unlocks after V2 Phase 2.