
Build a VC portfolio AI classifier
You'll have a working script that flags which of your portfolio companies face circular revenue risks vs. real customer demand.
What you'll make
Portfolio Revenue Risk Classifier
- load_portfolio_data() — data ingestion from your portfolio tracking: what to include
- score_revenue_risk() — the core classification logic: what to include
- generate_flags() — high-risk company identification: what to include
- export_results() — formatted output for investment decisions: what to include
Paste your complete Python script with test output showing risk scores for at least 5 companies
The 3 parts
Each part ends with a check you can see.
Map Portfolio Revenue Sources
Extract actual revenue sources from 10+ portfolio companies to find circular patterns
Done when
A spreadsheet with 10+ companies showing top 3 revenue sources each, marked as circular vs. independent
Build Risk Scoring Logic
Code a scoring function that weighs circular revenue indicators and test it on your mapped data
Done when
Working score_revenue_risk() function that produces 1-10 scores for all mapped companies with at least 3 different risk factors weighted
Test Classifier Against Known Cases
Run the complete classifier on your portfolio and validate results against 3-5 companies where you know the ground truth
Done when
Complete classifier tested on full portfolio with accuracy checked against at least 3 companies where you have definitive knowledge of revenue risk levels
Capstone: you put together your Portfolio Revenue Risk Classifier.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Run a working classifier on 10+ portfolio companies that flags circular revenue vs. genuine customer demand
- Identify the 3 most common revenue risk patterns in your AI portfolio companies
- Have a documented scoring framework you can apply to new deals during due diligence
- Know which financial signals AI gets right vs. where you need domain judgment
Who it's for
- VCs managing AI-heavy portfolios who need systematic risk assessment
- Investment professionals who can copy-paste code and modify scoring rules
- Portfolio managers tracking revenue quality across multiple AI companies
What it won't cover
- Building production-grade fund management systems
- Legal compliance for automated investment decisions
- Replacing human judgment in investment committee decisions
- Private company financial statement analysis fundamentals
Skills you'll practise
- Circular revenue dependency evaluation
- AI-assisted pattern recognition in financial data
- Investment thesis stress-testing with automation
- Risk signal calibration across portfolio companies