BarterNow

Event sponsorship in India runs on cold calls, WhatsApp groups and spreadsheets. Agencies take 15 to 40% for introductions, nobody can prove the deliverables happened, and both sides find out what went wrong after the money has moved.

One workspace for the whole deal: discover, match, contract, pay, deliver, prove. Brands see events scored on category, city, budget, audience and deliverables, each score with a plain-language reason. A deal moves into a negotiation room where every deliverable is voted on line by line, the agreed board becomes a contract both sides sign, and payment splits automatically: 90% to the organiser, a flat 10% to the platform.

After the event, organisers upload proof for each deliverable, brands approve or challenge it, and missed items turn into a proposed deduction instead of an argument. Brands also get tracked links, a conversion pixel and an AI-written summary of how the sponsorship performed.

Nine containers in production on AWS behind Cloudflare: a Next.js app with 129 pages, an Express API with 30 modules, a payments worker, an admin console, a FastAPI matchmaking service, a RAG support chatbot and nginx. PostgreSQL holds 111 models across three schemas, including a separate audit database.

Payouts run through a state machine with exponential retries and alerts. Bank and PAN details are encrypted with AES-256-GCM and masked in every response. Every pull request goes through security gates (dependency audits, secret scanning, smoke tests) before it deploys to EC2 with automatic rollback.

I lead sales as well as engineering: positioning, pitch decks, a cold-email system I built on AWS SES with AI-personalised openers, and the onboarding calls. BarterNow went live in April 2026, has 125+ organisations signed up, and is backed by a ₹1,00,000 TIDE 2.0 grant through MeitY. Early organisations include Extramarks, Contral AI, Legal Bhasha and Axiom Ayurveda. I also hired and ran three interns across AI, frontend and backend.