Leaving Cleo, Joining EF, Starting Fabric

On leaving a role I loved, searching for the right problem, and building something new

I left Cleo while exploring ideas in personalised consumer health. After hundreds of user interviews, I kept running into the same problem: I found many interesting pain points, but none I cared enough about to spend decades of my life on.

There were many problems worth solving just none that felt like my problem.

So, I moved on.

I joined Entrepreneurs First and spent the summer of 2024 working through different ideas with different co-founders.

At EF, I met Massimo and we started talking seriously about Article 6(9) of the EU’s new Digital Markets Act. We uncovered some deep secrets about context portability which led us to founding Fabric.

We moved to San Francisco (back in London for now) and raised $3M in weeks.

On a side note, if you’re a young founder, joining EF is exactly the kind of low-risk, high-reward decision worth taking.

It won’t guarantee success, raising millions of dollars after demo day is the exception not the norm, and raising a large round only buys you more time to find product market fit.

There’s very little downside if you don’t find your calling (or your co-founder) and enormous upside if you do.

Back to the main story. Article 6(9) of the Digital Markets Act mandates programmatic data portability for gatekeepers. In simple terms: if you’re a designated Big Tech company: Google, Meta, Amazon, LinkedIn, Booking, or TikTok, you are required to provide a user’s data (Google searches, Instagram stories, and so on) to any company the user chooses, in a real-time, programmatic way.

This was our game-changing insight.

We had seen this story before. For years, banks kept our financial data locked inside their systems. With no real competition, there was little incentive to innovate. That changed in the mid-2010s, when companies like Plaid and TrueLayer made it possible for users to share their data with consent. Almost overnight, new products like Cleo and Chime became possible, and an entire wave of consumer fintech followed.

The same tipping point was arriving for consumer preference data.

We are building your digital self. Something you can bring into your favourite AI products to unlock perfectly tailored experiences.

With Fabric, you bring rich personal context from your Instagram stories, Google searches, YouTube watch history and more into AI apps like Claude and ChatGPT. So your AIs stop being clueless agents and start feeling more like friends who actually get you.

We’re creating a new market, and the startup graveyard is full of companies that tried to do the same. The path forward isn’t laid out, we find it by running experiments, discovering what users actually respond to, and following those signals toward something that might become product-market fit.

This kind of work demands conviction more than certainty. Effort doesn’t translate cleanly into outcomes. When your company is closely tied to your sense of purpose, failed experiments have a way of colouring everything else. Progress rarely feels linear. More often, it’s a long sequence of attempts that only make sense in hindsight.

When we pivoted from personalising experiences for e-commerce brands to building directly for users, it felt like a personal failure. All those investor questions about go-to-market came back to me. But the lessons that actually change how you operate come from experiments that don’t work.

And yet, I wouldn’t trade it for anything. Working with a co-founder like Massimo, someone philosophically aligned with genuinely complementary skills, makes the uncertainty more navigable and the wins more shared.

Getting to work on a problem that is meaningful, intellectually demanding, and still unfolding is its own reward.

Last year, I was searching for a problem that felt like mine. I’m no longer searching.