Roman Licursi
GTM engineer building production AI agents, revenue systems, and automation. CS @ UW-Madison.
This profile was put together from Roman's public GitHub portfolio, and they haven't joined Vibetown yet, so reach them through their portfolio. Is this you? Contact us to update or remove it.
About
Roman Licursi works in AI GTM & Growth, AI automation & agents and AI engineering, based in Madison, WI (Together AI). GTM engineer building production AI agents, revenue systems, and automation. CS @ UW-Madison. Public projects on GitHub include Salesforce Lead Routing Pipeline (Salesforce-integrated lead scoring & routing pipeline: Python ML model scores leads via logistic regression,…), Lead Scoring Automation (Case study: rule-based and ML lead scoring with synthetic CRM data and Slack alerts; modeled a projected 12%…) and Agent Telegram Runner (Telegram completion pings for local Codex and Claude Code runs), mostly written in Python and TypeScript.
Projects 6
- github.com
Salesforce Lead Routing Pipeline
Salesforce-integrated lead scoring & routing pipeline: Python ML model scores leads via logistic regression, syncs scores to Salesforce CRM via REST API, and triggers native Record-Triggered Flows for Hot/Warm/Cold tiering and SDR task creation.
- github.com
Lead Scoring Automation
Case study: rule-based and ML lead scoring with synthetic CRM data and Slack alerts; modeled a projected 12% MQL-to-SQL lift.
- github.com
Agent Telegram Runner
Telegram completion pings for local Codex and Claude Code runs.
Research Podcast Factory
Local-first workflow for turning research topics into NotebookLM podcasts and Spotify episodes
github.com- github.com
PodBot---AI-Podcast-Discovery
Full-stack AI podcast discovery app built with React, TypeScript, Spotify OAuth, and Supabase.
Career Os Public
Six-layer AI career intelligence system — automated market research, synthesis, sprint cards, and outreach drafting via GitHub Actions + Claude
github.com