Show HN: Semantic search engine for finding relevant VCs and angels (graph.one)
Technical implementation:
- Indexed 9k active VC funds and ~5k angels globally
- Fine-tuned a custom embedding model on real-world pairs of investors and companies
- Developed an identity resolution system to collate portfolio data across many sources including portfolio websites scraped with a custom AI agent
Key features: - Search full investor database without signing up
- Optional email/calendar integration shows 1st/2nd degree connections to investors
- Results ranked by semantic similarity to historical investments
Example query: "B2B SaaS platform for automating accounts payable workflows using OCR and ML" → Returns VCs/angels who've invested in similar enterprise automation/fintech companies
Origin: Built this after struggling with my own seed round. Existing solutions didn't solve the core problem: systematically finding ALL relevant investors, not just the obvious ones.
Would appreciate feedback on:
1. Search result quality/relevance
2. Additional data points that would be useful
3. Performance/latency issues if you encounter any
Team: danishabdullah, combdn, thelifeofrishi, howes, Hossamhamdy20, Isomorpheus
6 comments
[ 2.9 ms ] story [ 26.5 ms ] threadWould be awesome if I could see a family tree graph with my connections aswell
But importantly: We don't get access to the email body/content