Client / Industry Information:
Client: Internal Product for
Company: Liquidity Capital
Role: AI Product Manager – Investment Intelligence
Duration: March 2023 – March 2024
Background:
Add paragraph text. Click “Edit Text” to update the font, size and more. To change and reuse text themes, go to Site Styles.
Problem Statement:
As Liquidity Capital scaled, our Sales and Investment teams were flooded with inbound leads, but:
Many startups were too early or didn't meet key financial or sector criteria
Investment teams spent hours vetting pitch decks manually
Founders often waited weeks to hear back—if at all
There was no standardized way to match a startup’s “funding cause” with relevant investor appetite
We needed a screening solution that balanced speed, accuracy, and investor relevance—without burdening internal teams.
Approach:
Discovery & Data Analysis
Interviewed Sales, Investor Relations, and Investment Analysts
Audited 200+ past startup applications to define key eligibility signals
Mapped recurring patterns in high-conversion deals (growth rates, sectors, geos)
Solution:
Front-End Intake: React app integrated with Hubspot CRM
Data Processing: Python, LangChain, OpenAI (GPT-4 for pitch and intent parsing)
Scoring & Matching Engine: Rule-based logic + LLM-assisted categorization
Data Storage & Search: PostgreSQL + Pinecone for thematic investor matching
Governance & Logging: Role-based routing, audit trails, feedback loop for corrections
Try the live demo
Key Outcomes / Results:
Reduced unqualified lead handoffs to Sales by 42%
Improved lead-to-meeting conversion rate by 23%
Delivered startup screening results in under 1 minute, down from 3–5 days
Enabled thematic routing to investor desks based on startup’s funding need
Created a reusable model for early-stage eligibility checking across new geographies