Vcomply

Real-time AI compliance, without the complexity.

Comment

Inspiration

AI regulation is no longer theoretical—it is actively impacting companies today. As laws emerge across jurisdictions, teams struggle to keep track of what applies to them, often relying on manual research or legal consultations. We've been hard at work with this idea for months.

We were building not just for a hackathon but as an ongoing startup, with the belief that compliance should evolve from a static, periodic process into a real-time system:
$$ \text{Compliance} \rightarrow \text{Continuous, System-Level Intelligence} $$

What it does

VComply is a regulatory intelligence platform that maps a company’s AI use cases and jurisdictions to applicable laws, identifies compliance risks, and provides actionable next steps.

It transforms fragmented legal requirements into a structured, live compliance layer:
$$ \text{Compliance State} = f(\text{Use Cases}, \text{Jurisdiction}, \text{Regulations}) $$

How we built it

We built a full-stack MVP designed to scale beyond the hackathon:

  • Frontend: Next.js + TypeScript
  • Backend: FastAPI (Python)
  • Database: Structured regulatory dataset
  • Core Engine: Rule-based applicability logic
  • AI Layer: Generates compliance summaries and insights

System pipeline:
$$ \text{Company Input} \rightarrow \text{Applicability Engine} \rightarrow \text{Risk Assessment} \rightarrow \text{Dashboard} $$

The focus was not just speed, but building a foundation for a real product.

Challenges we ran into

1. Structuring legal data

Regulations are not written for machines. Converting them into structured, queryable formats required abstraction and simplification.

2. Defining accurate applicability

Applicability depends on multiple dimensions:
$$ \text{Applicability} = f(\text{Jurisdiction}, \text{Use Case}, \text{Company Attributes}) $$
Capturing this correctly without overcomplicating the system was difficult.

3. Balancing MVP vs long-term vision

As an ongoing startup, we had to decide what to build now versus what to defer:

  • avoid overengineering
  • still maintain a scalable architecture

Accomplishments that we're proud of

  • Built a functional MVP with a clear end-to-end flow
  • Created a system that translates complex regulations into actionable outputs
  • Established a scalable architecture for future development
  • Positioned the product as more than a hackathon demo—an early-stage startup

What we learned

1. Compliance is fundamentally a data problem

At its core:
$$ \text{Compliance} = \text{Mapping} + \text{Evaluation} $$

2. Narrow scope drives clarity

Focusing on specific AI use cases and jurisdictions made the product significantly more effective.

3. Execution matters more than ideas

The value comes from building a working system, not just defining the concept.

What's next for VComply?

  • Expand regulatory coverage across more states and use cases
  • Automate regulatory data ingestion (e.g., monitoring updates)
  • Introduce continuous compliance monitoring.
  • Build integrations with HR and AI systems
  • Validate with real customers and iterate toward product-market fit

Long-term vision:
$$ \text{Static Compliance Tools} \rightarrow \text{Real-Time Regulatory Infrastructure} $$

Built With

Share this project:

Updates