SANTRA AI - GTM Intelligence Agent

Autonomous GTM intelligence for B2B teams — monitor competitors, collect live evidence, and approve CRM actions before anything runs.

Comment

Inspiration

B2B revenue and competitive intelligence teams spend 15–30+ hours per week manually checking rival pricing pages, job boards, SERP rankings, and product changelogs. Signals get missed, battlecards go stale, and leadership resists automation because nobody wants a CRM workflow firing on unverified intelligence.

We built SANTRA AI to close that loop: an autonomous agent that watches competitors continuously, collects live web evidence, synthesizes executive-ready briefs, and proposes CRM actions — but only executes after a human approves.

What it does

SANTRA AI is an autonomous B2B GTM intelligence agent with four core modules:

  1. GTM Competitive Monitors — Describe what to watch in plain language (pricing, hiring, launches, SERP moves). SANTRA runs a full agent loop: intent parsing → dynamic tool routing → evidence collection → change detection → executive synthesis.
  2. Competitor Intelligence — Deep-dive analyst workspace for battlecards, positioning gaps, and differentiation playbooks.
  3. GTM Command Center — Turn signals into ICP, channels, launch plans, and growth strategy.
  4. GTM Advisor — Interactive chat for competitor questions, grounded in live web evidence.

Human-in-the-loop by design: When a monitor detects a material change, SANTRA queues a proposed CRM/webhook action. Nothing fires until a user clicks Approve in the Action Queue.

Example monitors:

  • "Alert me when [Competitor] changes enterprise pricing tiers"
  • "Track competitor hiring in enterprise sales roles"
  • "Monitor SERP rankings for our category keywords"

How we built it

Stack: Next.js 15 · React 19 · TypeScript · Supabase · Tailwind CSS · Framer Motion

Agent architecture:

Built With

  • ai
  • artificial-intelligence
  • automation
  • b2b
  • bright-data
  • cloudflare
  • competitive-intelligence
  • framer-motion
  • gtm
  • human-in-the-loop
  • machine-learning
  • mcp
  • netlify
  • nextjs
  • openai
  • playwright
  • react
  • revops
  • saas
  • supabase
  • tailwindcss
  • typescript
  • vitest
  • web-scraping
  • webhooks
Share this project:

Updates

posted an update —

I worked on the frontend, building out the UI with Next.js, TypeScript, and Tailwind CSS, and adding the Framer Motion animations for a polished feel. I also handled testing/QA using Vitest and Playwright to make sure the agent workflows and UI held up under real usage. Great experience working across both design and reliability for the product.

Log in or sign up for Devpost to join the conversation.

Submission history