Building the AI Brain of Wealth Management

The Brand Called You is a show about journeys, and host Ashutosh Garg spends the first third of this episode drawing out Vineet Mohan's before he lets him anywhere near a product pitch. It's the right call. Mohan's fourteen years at HSBC — moving every two or three years to a new leadership role in a new part of the world, from India to the US, the UK, Scotland, and Hong Kong — turns out to be the argument for why FastTrackr exists at all.

Mohan is the co-founder and CEO of FastTrackr AI, an AI-led platform purpose-built for the US wealth-management industry. But the through-line of the conversation is that a vertical AI company is only as good as the operator behind it, and this is a portrait of that operator.

What Edinburgh in lockdown taught him

Asked which experience shaped him most, Mohan doesn't hesitate for long before landing on Scotland. He and his family moved from the US to Edinburgh in the middle of COVID with a one-year-old, into a city where they knew no one, facing a six-month lockdown. Personally it was hard. Professionally it was the most demanding thing he'd done. HSBC's European payments business had been decentralized, with client-management roles scattered across markets, and his remit was to build a brand-new function to centralize the whole European operation — onshore in Edinburgh, offshore in India — which meant hiring hundreds of people.

You'd want to do that in person, so culture sets from day one. He couldn't. Everything was on Zoom. So he and his leadership team made a deliberate choice: instead of hiring only for financial-services experience, they brought in people from different countries, languages, and professional backgrounds. Within five or six months the team was engaging thousands of clients daily.

"It was a massive experience in how diversity helps when you run a business, and how adversity is something you can definitely overcome if you have the right people and the right attitude."

That instinct — that the ones building something from the ground up gave him the most satisfaction — is what he traces the founder itch back to. He'd wanted to build since growing up in 1990s India, started angel investing a decade ago to be part of the startup ecosystem vicariously, and realized the roles he'd loved at HSBC were all the stand-it-up ones: the FinTech partnerships business on the US West Coast, the bank's first global corporate-banking learning academy. When he moved back to India after HSBC, building made sense.

A massive industry that stayed manual

Garg asks the core question directly: what problem in US wealth management did he set out to solve? Mohan's answer is a tour of the industry's structure. It's vast — the wealth generated and accumulated after the world wars matured into a booming advice industry, now hundreds of trillions under management and close to a million people employed. And it's fragmented, from 30,000 independent shops to broker-dealers to large banks, each serving clients differently. The tech is fragmented too: wealth management is more than investments — tax strategy, estate planning, education planning — and a different piece of niche software grew up around each, so a typical firm runs five to 20 tools that don't talk to each other.

Underneath the fragmentation is the part that drew him in. As they peeled the onion, he says, they found the industry was manual in almost everything — managing meetings, analyzing documents, collecting and validating and entering data. Why had that gone underserved for so long, given the money in the industry? Because the technology wasn't ripe. Earlier automation like RPA was rule-based; it couldn't understand context and take action.

"A lot of the systems in use are systems of record rather than systems of action. You'll have a CRM, a financial planning tool, a portfolio management tool, all housing slightly different data about your client — and humans have to keep them in sync."

A pre-filled form instead of a blank one

When Garg asks him to walk through a real day, Mohan describes an RIA anywhere in the country. Before a week of client meetings, FastTrackr prepares nuanced briefings on who the advisor is meeting based on past interactions; afterward it automates the follow-up — notes, action items, a drafted email. Then there's the document workload: brokerage statements and tax forms that someone would otherwise read page by page to extract and enrich. Upload them and the platform recognizes the document type, pulls the data, and helps generate proposals and insights.

But the use case he clearly cares about most is onboarding. The usual flow is backwards — you talk to a prospect several times, shake hands, and only then send a blank intake form asking for 120 data points. If the AI has been in those conversations and received documents along the way, it starts enriching a master record from the beginning, so onboarding means asking the client to validate what's there and fill what's missing, then flowing that data into account-opening forms and the rest of the stack. Asked what advisors automate first, he's honest that the meeting assistant is the obvious answer — everyone can see and believe it because they've used ChatGPT in their own lives — but the deeper value is in the unstructured-data and data-gathering work where teams lose 25 to 30% of their week.

There's a texture to how firms of different sizes respond, too. Mohan's read is that whoever invests in technology in a structured way — not just AI — tends to see more growth in AUM, the industry's key metric, than firms that don't. Small firms decide fast: it's a founder-owner who says yes or no depending on where they sit on the technology curve, and once they say yes they can start with a single module and expand from there. Larger firms engage more deeply, with in-house teams around security and compliance asking far more detailed questions, so the sales cycle runs longer — but volume and scale make up for it over time.

Trust is the real barrier

Garg presses on the questions that matter in a regulated industry: accuracy, auditability, compliance. Mohan calls it "the big question" and tells a story that lands it — an RIA leader who reached out not to hear the pitch but with a list of compliance questions, saying he'd only engage if every box was ticked. Fair enough, Mohan says, when people are discussing sensitive topics and sharing private information. Client data is never used to train models; everything is encrypted in transit and at rest; access is tightly controlled. Coming from banking, he says, made the importance of that obvious from day one. He also mentions the eight-point checklist from an earlier webinar — eight questions any advisor can put to a technology vendor — which resonated because so many leaders had asked him what they should even be asking.

The subtler barriers are human. With a median advisor age around 55, tech awareness can't be assumed, and the industry still runs on word of mouth — for winning clients and for choosing vendors. So an early-stage company has to earn trust through industry events, warm networks, and association partnerships; being able to say "I'm used by several of your peers" changes the conversation.

On what success looks like, Mohan wants FastTrackr to be seen as an extension of the team — hand the repetitive, error-prone grunt work to the platform rather than to another hire — which means becoming comprehensive enough to cover what advisory teams actually do. He wants a couple of new modules live as the year turns, and expects the numbers — users, market share — to follow the coverage rather than lead it.

His closing advice to corporate leaders eyeing an AI venture is the line the episode keeps circling back to: 40 is the new 20. Accelerators back fearless 25-year-olds, and always will, but vertical AI rewards people who've lived a workflow's edge cases for fifteen or twenty years. Those edge cases, he stresses, are massive in their own right, and someone who has seen a workflow every day for a decade or two can now realistically build something that reshapes it. The tooling to spin up a proof of concept is finally cheap and fast, often without writing it yourself, and there are plenty of people who can help build the rest — so the opportunity is more accessible than it's ever been. Garg, delighted, mentions the AI "digital twin" a group of friends built from his own writings, now popular in its own right, before landing the episode on Mohan's steady closing note: the human stays firmly in the loop, and there is still enormous room for it.