Intern Project
Julia Zheng
Duke University • Sophomore
Regulatory & Policy Exposure Intelligence Engine
The Regulatory & Policy Exposure Intelligence Engine is an AI-powered diligence tool built for Visible Ventures to evaluate regulatory risk across early-stage startup deal flow. Given a startup submission, including company name, sector, geography, revenue model, and stage, the system autonomously synthesizes regulatory context and generates a structured, decision-ready risk assessment within seconds.
The workflow is orchestrated in n8n and leverages a multi-step AI pipeline to transform unstructured inputs into standardized outputs. It begins by enriching the company profile with relevant regulatory context using web-based research via SerpAPI, focusing on sector-specific policies, governing bodies, and compliance requirements. This information is then processed through a sequence of LLM-powered agents (Groq-hosted LLaMA models), which extract key signals and map them into a structured regulatory scoring framework.
The scoring engine evaluates companies across multiple dimensions of regulatory exposure, producing a Regulatory Exposure Score (1–10) based on overall compliance burden and downside risk, a Policy Volatility Risk rating (Low / Medium / High / Critical) capturing sensitivity to regulatory change, and a Regulatory Durability rating (Fragile → Fortress) reflecting how defensible the company is to long-term policy shifts. These outputs are synthesized into an AI-generated executive memo, which includes regulatory tailwinds, headwinds, and targeted diligence questions for investors.
For companies exceeding a predefined threshold (Exposure Score ≥7), the system triggers a deep regulatory risk analysis module, which models specific failure scenarios (e.g., HIPAA violations, SOC 2 non-compliance, state-level licensing gaps), estimates potential financial and operational impact, identifies regulatory triggers to monitor, and proposes mitigation strategies. The final outputs are automatically logged to Google Sheets for pipeline tracking and delivered via email to the analyst.
Regulatory & Policy Exposure Intelligence Engine — UI Images
Inputs
- Company name
- sector (e.g., Healthcare, Fintech)
- geography (e.g., United States)
- revenue model
- stage (e.g., Seed, Series B)
Outputs
- Regulatory Exposure Score (1–10)
- Policy Volatility Risk classification (Low/Medium/High/Critical)
- Regulatory Durability rating (Fragile → Fortress)
- AI-generated executive memo with regulatory tailwinds, headwinds, and diligence questions
- deep risk analysis for high-risk companies (Exposure Score ≥7) with failure scenarios and mitigation strategies
- Google Sheets entry (structured log)
- email alert to analyst
About Julia Zheng
Julia Zheng is a sophomore at Duke University studying Computer Science and Economics (Class of 2028). At Visible Ventures, she built an AI-powered regulatory and policy exposure intelligence engine that helps investors evaluate regulatory risk in early-stage startups, bringing structure and consistency to one of the most fragmented parts of venture diligence.
“Building this project changed how I think about diligence. It helped me better understand how venture investors approach early-stage decisions, especially in areas like regulatory risk that are often overlooked or inconsistent. Using tools like n8n and LLMs, I was able to turn that ambiguity into a more structured system, and see how AI can bring more clarity and consistency to those decisions.”
— Julia Zheng