LinkedIn
Devesh is an AI product manager who thrives in chaos, the kind of calm that isn't indifference, but clarity. Born in Varanasi, one of the world's oldest living cities, he carries with him a rare ability to hold many perspectives at once, a skill that turns out to be surprisingly useful when you're building things for people.
His path to product management career was anything but straight. He started in Sales, far from roadmaps and sprint reviews until, almost by accident. He found himself managing a handful of digital products in a role that hadn't quite been defined yet. Something clicked. In the quest of the transition to product management, he pursued a management degree, which gave him the frameworks to think about strategy, markets, and business outcomes. But the instinct? That came earlier, from a career that taught him how people actually make decisions.
Today, he is a Product Manager at Tredence, Bengaluru, and the driving force behind TReK 2.0, the organisation’s first GenAI-powered enterprise knowledge platform. His journey is proof that geography is never a ceiling, and that ownership is earned, not granted by a job title or years of experience.
In conversation with us, Devesh candidly share what a day in his life as a product manager, working on enterprise AI platforms looks like.
How Growing Up in Varanasi Shaped My Perspective
Varanasi is a city that teaches you patience and perspective. As an ancient city with a rich culture, things move there at a slow, and steady pace, and yet everything feels deeply meaningful. Growing up there, I did not have the same access to tech communities or startup culture that someone in Bengaluru or Mumbai might have had. But I had something else: a deep curiosity, a comfort with ambiguity, and a stubborn willingness to figure things out on my own.
When I moved to Bengaluru for engineering, I was navigating a new city, a new culture, and an entirely new pace of life. That experience of being the outsider who had to earn his place never left me. I think it made me a better product manager, who can work on AI product development. You learn to listen carefully before you speak, ask better questions, and never walk into a room assuming you already have the answer.
A Non-Linear Path into Product Management. Can you explain more on this?
Would be cliched to say it was destiny. I’d say, I architected this path after understanding my love for delivering smooth user experience. From a career in sales, on the ground, talking to customers every day, understanding why people made the decisions they made, I gradually made the switch to working for AI in enterprise applications. My sales experience was invaluable. It built in me a reflex I still carry: always start with the human, not the feature.
After that, I spent time managing digital products, owning lifecycle from idea through launch, measuring what worked, and learning to make fast decisions with incomplete information. From there I pursued a management degree, which gave me the frameworks to think about strategy, markets, and business outcomes.
By the time I joined Tredence in 2022, I had a layered foundation: commercial empathy from sales, strategic thinking from my MBA, and hands-on product execution. No single piece of that was the “right” background for enterprise AI product management. All of it together was.
Why Tredence?
What excited me about Tredence enterprise AI platforms were built with a lot of care. Every product here had a direct, measurable impact on how large enterprises run. That rigour appealed to me.
I started by owning the roadmap for a suite of data engineering accelerators, tools that helped Fortune 500 organisations assess migration readiness, trace data lineage, and rationalise reporting assets at scale. These ran across 20+ client environments and were used by hundreds of data engineers.
Those early years taught me how to operate in a matrixed organisation, align engineering, data science, and business teams around a single product vision, and translate deep technical complexity into something a senior stakeholder could act on. That muscle is everything in enterprise product management.
Building TReK 2.0: An AI Product Manager’s Approach
TReK 2.0 was born out of a very real, very human frustration: people at Tredence, brilliant consultants, analysts, and engineers, were losing enormous amounts of time just searching for information that already existed somewhere in the organisation.
A proposal written six months ago. A policy document buried in a shared drive. An expert who had solved the exact same client problem last quarter. That knowledge existed; but it was scattered, siloed, and effectively inaccessible to the people who needed it most.
TReK 2.0 is our answer to that. It is a GenAI-powered enterprise knowledge platform that brings organisational intelligence into a single, secure, conversational interface. The impact has been tangible: 50% organisation-wide adoption within nine months, and query resolution time reduced by 30–40% as employees shifted from manual searching to intelligent self-service.
“The number I remember most is the consultant who used to spend one or two days hunting for a case study. After TReK, she found it in under thirty seconds. That is not a metric. That is two days of her life returned to her.”
A Day in the Life of a, AI Product Manager
Daily Responsibilities of an AI Product Manager in Enterprise AI
This is the question I love most, because I think the PM role is deeply misunderstood; especially by people who have not seen it up close. My day has roughly three rhythms to it, and I have come to think of them as three different jobs packed into one.
First Half: Enabling the Team
The first part of my day belongs to the engineering and design team. I am in conversations with data engineers, data scientists, and the UI/UX tea, enabling them, walking through acceptance criteria, helping them understand the intent behind a feature rather than just its specification. Great engineers and designers can build anything. My job is to make sure they are building the right thing. That means being precise enough in how I communicate that there is no ambiguity, and available enough that when a question surfaces, it does not become a blocker.
Second Half: Discovery with Stakeholders
The second part of the day is focused on the business stakeholders. I spend significant time with the stakeholders and end users of TReK: employees across Tredence. This is discovery work, and it is the part of the job I am most passionate about.
I sit with them and map their current journey in detail. Where are the friction points? What are the three steps they take every time they need a piece of information? Where do they give up and just ask a colleague instead? I call this understanding the as-is journey, not just the surface-level complaint, but the underlying pain.
From that, I build the to-be journey: a picture of how the same process should feel once TReK addresses those pain points. And then I translate all of that into user stories that go straight into the product backlog. It is where the product actually gets shaped.
Finally, Always on Research and Staying Ahead
My most favourite part of the day is research. I follow what is happening in the GenAI space closely, new interaction patterns, emerging product paradigms, UX shifts in how users engage with AI-powered interfaces. My job is to bring the best of what is happening externally into what we are building internally. TReK should not just be good by today’s standards. It should be ready for where things are heading.
Stakeholder Management and Strategy. What does that look like and why does it matter?
Sometimes daily, and sometimes, mostly on a weekly basis, I am in conversations with Directors and VPs, presenting roadmap updates, defending prioritisation decisions, and securing buy-in for the next quarter’s features. That visibility is both a privilege and a responsibility. We have an open floor culture, where everyone is encouraged to share ideas and innovate.
I own the quarterly feature plan for TReK. That means I am not just executing a roadmap someone else defined, I am the one making the case for what gets built, in what order, and why. Every quarter, I walk into a room with senior leadership and explain the logic: here is what we learned, here is where users are struggling, here is what we are going to build, and here is why I believe it will move the needle on adoption and impact.
“At Tredence, ownership is not something you wait to be given. If you can do the work and make the case, you will be trusted to lead it.”
Product Ownership in AI Product Management: From Strategy to Execution
It means you are accountable for the outcome, not just the output. Anyone can write requirements or fill a backlog. End-to-end ownership means you care whether the feature actually gets adopted. You track retention. You watch usage data and ask why something is underperforming. You go back to users and find out. And then you iterate.
I own the success metrics for TReK, adoption, engagement, reliability, and how the platform is actually changing how people work. If a feature ships and users are not engaging with it, I do not move on. I go back, understand where the discovery process broke down, and fix it.
That is what makes product management at Tredence genuinely different from being a project manager with a fancier title. The bar is outcomes. And the organisation holds you to that bar while giving you the latitude to pursue it your way.
Winning the Process Excellence Champion Award in November 2025 was my crowning glory. This recognition was not just for what I delivered, but for how I worked: the rigour, the cross-functional alignment, the investment in making the processes around me better for everyone.
Coming from Varanasi, where I was often the person who had to prove himself in rooms that were not built with him in mind, those recognitions matter. They are a reminder that your background is not your ceiling. Your work is.
How to Build a Career in AI Product Management
Career Advice for AI Product Managers from Tier-2 Cities
Do not wait until you feel ready. You never will.
The AI space is moving faster than any pedigree or credential can keep up with. Nobody has a ten-year head start on what is being built today. The person who will create the most impactful AI products of the next decade might be in Varanasi right now. Or Sikkim. Or Indore. Or Coimbatore. Geography is not the variable that matters.
What matters is whether you are genuinely curious. Whether you are willing to sit with a stakeholder and really listen, not to confirm what you already think, but to be changed by what they tell you. Whether you can hold ambiguity without panicking and turn incomplete information into a clear decision. Whether you are willing to unlearn what worked last year because the landscape has already shifted.
At Tredence, those qualities are what get you in the room, and what keep you there. Where a PM who took a winding road, from a tier-2 city and a non-linear career, can end up owning the product vision for the company’s most ambitious platform, presenting quarterly roadmaps to leadership, and building something that genuinely changes how people work.
That opportunity is real. And it is available to anyone willing to earn it. Explore careers at Tredence.
