Co-founded a platform that meets founders and investors at whatever stage they are already at, built on a verified-network moat: proof-of-work founders, verified investors, and crowd-validated problem signal, with adversarial AI review acting as a trust filter rather than the product.
Case Study / Co-founded platform
Something
Team formation and capital discovery are the same failure wearing two costumes. Existing platforms either gate on post-traction signals a pre-seed founder cannot produce, or apply no verification at all and drown both sides in noise.
Next.js 15, React 19, and Radix UI frontend over a Node.js and Express backend with MongoDB, using LangGraph orchestration and cost-aware multi-provider AI routing across Groq, Gemini, and Claude.
573-person waitlist inside the first month with zero paid distribution. I own frontend, AI architecture, and go-to-market as co-founder and CEO.
Technical highlights
What to inspect.
01
Verification gated on proof that exists before revenue (prototype, pilot, patent, press) rather than on Stripe-verified revenue or accredited-investor status.
02
Private review track where AI reviewers argue for and against an idea before anything becomes public.
03
Cost-aware routing across three model providers, chosen per request against latency and price rather than pinned to one vendor.