From Idea to MVP: Building a Compliance-First AI Tool in Boston’s Startup Scene

If you had asked me two years ago whether I’d be building an AI startup focused on IRB and research compliance, I would’ve probably laughed. Like many founders, I didn’t intentionally walk into this space. I simply kept running into the exact same pain point over and over again - and eventually, the problem became too loud to ignore.

I spent years in academic and medical research across Harvard Medical School, MGH, Mount Sinai, and Northwell Health. No matter the institution or department, one thing was consistent: researchers were drowning in paperwork, emails, version control issues, and slow IRB approvals. Compliance officers were overwhelmed. Researchers were frustrated. And the workflow behind some of the most important scientific discoveries was being held together by PDFs and tribal knowledge.

That was the spark.

This is the story of how that spark turned into our MVP - a compliance-first AI tool - built from Boston, in the center of one of the world’s most complex research ecosystems.

Where the Idea Really Came From

My idea didn’t hit me as a sudden “aha moment.” It surfaced slowly, after years of sitting inside labs and watching colleagues struggle to complete forms, respond to reviewers, and interpret guidelines written in dense legal language.

Everyone had the same questions: “Which sections do I need to complete?” “Is this considered PHI?” “How do I know if this qualifies as expedited?” “Who do I email about this clause?”

And worst of all: “Why does this take so long?”

I realized something: researchers weren’t struggling because they lacked intelligence - they were struggling because the process wasn’t designed for them. It was designed for

compliance.

Why “Compliance-First” — Not “AI-First”

When AI hype exploded, many teams rushed to build “AI tools” first and figured out the real problem later.

We approached it the opposite way.

From Day 1, we built with one philosophy: Accuracy matters more than flash.

Our product isn’t entertainment. It’s not a chatbot for fun. It’s dealing with IRB policies, DUA agreements, PHI, data classifications, consent logic, and regulatory thresholds. If the system makes something up, it’s not an inconvenience - it’s a risk.

So instead of building a generic assistant, we built a compliance-first engine:

• It references actual IRB policy sections.

• It grounds every suggestion in real rules.

• It avoids duplicate or irrelevant questions.

• It behaves like a guide, not a black box.

This approach shaped everything that came next.

Why Boston Is the Hardest (and Best) Place to Build This

Boston is unique. It’s home to some of the most sophisticated research institutions and biotech startups on the planet. Everyone here understands the value of compliance - but also the pain of doing it manually.

Being in Boston meant I could talk to IRB coordinators, researchers, founders, and advisors who’ve lived this process for decades. Conversations were honest, at times blunt, but incredibly helpful. Some advisors even went deep into our prompts, UX flows, and policy reasoning.

That doesn’t happen in a vacuum. That happens in Boston.

The Hardest Parts Nobody Talks About

Every founder journey has its quiet battles. Here are mine:

1. Accuracy in a High-Stakes Domain

Compliance tools don’t get to hallucinate.We are investing huge effort into grounding the model with policy references and structured fields.

2. Balancing Innovation With Responsibility

In highly regulated sectors, you can’t “move fast and break things.” Each feature requires thoughtful design, not shortcuts.

But these constraints make the product stronger.

What Boston’s Ecosystem Taught Me

Three lessons stand out:

1. Talk to users early and often.

Whether they’re researchers, IRB officers, or founders - everyone wants this solved. People shared their stories freely, and those stories shaped our roadmap more than any internal brainstorming.

2. Advisors are invaluable.

One advisor fixed our entire question architecture with a single insight.Another helped us understand why certain features (like training verification automation) weren’t worth building yet.

3. Don't chase everything at once.

Boston is full of opportunities - accelerators, pilots, advisors, programs.The challenge is staying focused on the core MVP and what actually moves the needle.

A Note to Fellow Founders

If there’s one lesson I’d leave you with, it’s this:

You don’t need the perfect product to begin. You just need a painful problem, a few real users, and the commitment to improve every single week.

Our MVP wasn’t glamorous. It wasn’t huge. But it was real, and it solved real pain.

Boston gave us the problem, the pressure, and the people.

What’s Next for Us

What comes next for us is raising our first round of external capital. We’ve been bootstrapping the product for months, validating the problem across researchers, IRB teams, and early biotech founders, and now we’re ready to scale development and formalize our design partnerships.

Funding will allow us to expand our engineering capacity, deepen our compliance intelligence, and accelerate pilot programs - and we’re actively seeking investors who understand the pain and opportunity in research operations.

Our mission is simple: Reduce the friction, confusion, and delays in research compliance so science can move faster.

If you’re interested in following our journey, you can find us at labopshub.com or email me at ratnam (at) labopshub (dot) com.

About the Author: Ratnam Srivastava is the founder of LabOps, an AI platform simplifying IRB and research compliance workflows. Before starting the company, he spent five years in academic and medical research across Harvard-MGH, Mount Sinai, and Northwell Health, where he saw firsthand how administrative bottlenecks slow down science. He is now building compliance-first tools that help researchers and reviewers work faster with clarity and confidence.