Inspiration

We saw friends building GenAI tools for VCs on top of poor or outdated data—and it didn’t work. No Claude, ChatGPT, or LLM can do much without the right data. We realized the real problem: lack of real-time, high-quality, contextual data. That’s where Perplexity and Sonar step in.

VCs spend weeks analyzing startups, while founders struggle to find the right investors. Traditional due diligence is slow, manual, and misses critical insights. We wanted to fix this—with a "Tinder for VCs" that uses AI to match startups and investors instantly through deep, data-driven insights.

What it does

IntersectX is an AI-powered platform that analyzes pitch decks, founding teams, market claims, and competition. It generates investor-grade memos in minutes, not weeks. It matches startups with VCs using compatibility scores based on thesis fit, risk appetite, and stage focus—plus sends personalized market updates.

How we built it

We used FastAPI for the backend, MongoDB for storage, and integrated APIs like LinkedIn, GitHub, and market data sources. Our modular services handle document parsing, team analysis, market mapping, and risk scoring. Real-time chat and file processing are powered by AI engines. We used Agno as our LLM framework—flexible and abstracted for speed.

Challenges we ran into

API integration was tough—rate limits, data inconsistency, and formatting quirks. Accurate document processing across file types took effort. Building a smart matching engine that balances team, market, and risk factors needed multiple iterations. Public company info was slow to pull, so we built smart caching and parallel processing—still improving.

Accomplishments we're proud of

We’re a team of backend engineers, but in just four intense days, we built both the backend and a full-featured AI-powered UI from scratch. We created a robust analysis engine that ingests and processes multi-source data in real-time. The platform scales well and delivers investor-grade insights through an intuitive interface. We’re especially proud of how we blended traditional due diligence with cutting-edge AI to streamline the entire venture analysis workflow.

What we learned

Data quality beats data quantity—always. VCs care as much about confidence scores and data sources as the insights themselves. And real-time collaboration tools aren’t just nice to have—they’re essential for fast, confident decisions.

What's next for IntersectX

We’re adding patent analysis, technical feasibility checks, and regulatory compliance insights. Next up: predictive success models, video pitch + sentiment analysis, and a marketplace for expert diligence reviewers. Mobile apps are on the way—for VC insights on the go.

Built With

Share this project:

Updates