The Roundtable

How It Works

Three parts. One hour. No slides.

1

2 min

Quick Intros

Everyone at the table shares who they are, what they do, and what brought them to this topic — in under a minute each. No long bios. Just enough context so you know who you're learning from.

Think: the way you'd introduce yourself at dinner, not on a conference panel.

2

45 min

The Problem, The Scope, The Different Approaches

This is the heart of the session. We pick one real problem and go deep — not "what should we do about data quality" but "here's what we actually tried, here's what worked, here's where it broke." Both sides share openly.

From the Enterprise Side

Leaders share how they're tackling this problem today — what tools they picked, how they set things up, where they're still stuck. Real examples, not textbook answers.

From the Startup Side

Founders explain how their approach is different — what they're building, why they built it that way, and what they're learning from companies using it.

Think: "here's what actually happened when we tried this" — not "here's what the analyst report says."

3

13 min

Wrap-Up, Q&A, and One Thing to Try

We step back and connect the dots — what patterns showed up? Where do the approaches agree or disagree? Then the floor opens for any direct questions. Everyone leaves with one concrete thing they'll try before the next session.

Think: nobody walks away without at least one idea they can act on this week.

Who's Behind This

Built by people who've done the work, not just talked about it

Co-Founder · Governance & Foundations

Nikunsh Desai

Data & AI Strategy Leader

12+ years spent inside large enterprises — building the data rules, fixing the data gaps, and making sure AI projects had solid ground to stand on. Has worked with teams at Goldman Sachs, Morgan Stanley, UBS, and Marsh McLennan. Now helps data leaders figure out what needs to be true about their data before they invest in AI.

Marsh McLennan Goldman Sachs Morgan Stanley UBS

LinkedIn →

Co-Founder · Product & Adoption

Parul Verma

Data Product Management Leader

Product and data leader with nearly 12+ years at Nike's Enterprise Data & Analytics organization — governing messy data, building trust in it, and getting global teams to actually use it. Built enterprise data products across consumer, member, product creation, and supply chain domains at Nike scale. Brings the real-world perspective on what breaks, what works, and what it takes to make data AI-ready.

Nike Data Product Strategy Independent Advisor

LinkedIn →

Season One · Five Sessions

Data Foundations for AI Readiness

Every session tackles one version of the same question: what needs to be true about your data before AI can actually work?

01

Semantics + AI

If Your AI Doesn't Know What 'Revenue' Means, Nothing Else Matters

When one dashboard says you have 10,000 customers and another says 12,000 — your AI will get it wrong too. We explore how companies are fixing this.

The Opening QuestionOnly 7% of enterprises say their data is truly ready for AI. The biggest reason? Nobody agrees on what basic terms like "customer" or "revenue" actually mean across systems.

The Bottom LineYour data needs to mean the same thing everywhere.

↓ Tap for details

02

AI Governance

Who Governs the AI That Governs Your Data?

Every team is already using AI tools — often without anyone's permission. The question is: are you building guardrails that help people move faster, or ones that just slow them down?

The Opening QuestionYour marketing team is using ChatGPT. Your engineers are using Copilot. Your analysts are using Claude. Who's making sure all of this is safe, consistent, and actually helping?

The Bottom LineYour AI needs guardrails that speed things up, not slow them down.

↓ Tap for details

03

Data Quality

Your AI Is Only as Smart as Your Dirtiest Table

You can have the most advanced AI model in the world — but if you're feeding it messy, outdated, or inconsistent data, it'll give you confident wrong answers.

The Opening Question80% of enterprise AI spending didn't deliver the expected results last year. The models worked fine — the data underneath them was the problem.

The Bottom LineYour data needs to be fit for AI, not just clean enough for a dashboard.

↓ Tap for details

04

Ownership & Accountability

The Most Expensive Question in Enterprise Data: 'Who Owns This?'

You can buy the best tools in the world. But if nobody's name is next to a data domain, those tools are just expensive empty shelves.

The Opening QuestionA VP asks "why is this number different from last month's report?" and three teams point at each other. Sound familiar? That's an ownership problem, not a technology problem.

The Bottom LineSomeone's name needs to be next to every data domain.

↓ Tap for details

05

Data as a Product

Data Products That Ship vs. Data Products That Sit

Your data team built 47 dashboards last quarter. How many does the business actually open every week? We dig into what separates data products people use from ones they ignore.

The Opening QuestionIf your data product doesn't solve a problem that a real person has on a real Tuesday afternoon, it's not a product — it's a project that shipped and got forgotten.

The Bottom LineYour data needs to be built as a product, not a byproduct.

↓ Tap for details

Ground Rules

Simple rules that make honest conversation possible

🤝

Share Freely, Credit Privately

You can take ideas home and use them. You just don't say who said what. This is how we keep the conversation real.

🚫

No Sales, No Pitches

Nobody's selling anything in this room. No slides, no demos. Just people sharing what they've learned.

🔒

No Company Secrets

We talk about approaches, not proprietary data. Share the "how" — not anything that would make your legal team nervous.

🎯

Small on Purpose

3–5 people per table. We pick people who'll genuinely add to the conversation, not just listen.

Why Join

What you actually get out of this

If You're a Data Leader

See how startups are solving the exact problems you're struggling with — explained simply, without a sales pitch

Hear what leaders at other companies actually tried — including what didn't work and what they'd do differently

Leave with practical ideas you can bring back to your team this week

If You're a Startup Founder

Hear how enterprise buyers actually talk about the problem your product solves — in their words, not yours

Learn what's really blocking adoption — the stuff that doesn't come up in a sales call

Build real relationships with senior data leaders through shared learning, not cold outreach

Ready to pull up a chair?