Role: GTM Engineer
AI automation · Outbound infrastructure · CRM integration
Overview
A B2B sales team needed an automated, AI-driven outbound pipeline to replace manual lead research and end generic cold outreach that wasn’t converting.
Impact
The team was spending 20+ hours a week manually scraping prospect data and writing emails that returned under a 2% response rate. Hitting revenue targets meant personalisation at a scale no human could sustain, and freeing the team to take calls instead of doing research.
Work
I built an end-to-end “Invisible SDR” on Make.com with a frontier LLM at its core.
The pipeline identifies target B2B companies and enriches founder data through Clay, then passes that context to Claude with a prompt that produces a three-sentence pitch grounded in specifics about that company, not a template with a merge field. The drafted email syncs straight into HubSpot for one-click human approval before it sends.
The human stays in the loop for judgment and leaves the research to the machine. That’s the whole design principle: automate the labour, keep the discretion.
90% Reduction in Research Time
Eliminated manual scraping and data entry, returning 20+ hours a week to high-leverage sales calls.
One-Click Approval Workflow
Drafted, personalised outreach lands directly in HubSpot for human review, automation without losing control of what goes out under your name.
Context-Aware Personalization
Leveraged advanced AI to generate highly specific, account-based outreach at scale, significantly increasing positive outbound response rates.