Architecting India’s AI Future, One Agent at a Time
Nilesh Potdar
Chief Technology Officer
ResAI
Architecting India’s AI Future, One Agent at a Time
Nilesh Potdar
Chief Technology Officer
ResAI
In a tech landscape often obsessed with hyperbole, Nilesh Potdar quietly stands apart while engineering the future of AI with the same calm confidence that shaped his rise from rural India to global boardrooms. His journey, from a student grappling with low English proficiency to leading engineering at Microsoft, Amazon, and Goldman Sachs, is inspirational. It’s proof that rigour, resilience, and reinvention still define long-term success. Today, Nilesh is at the helm of three AI ventures, JupiterBrains, CanbizAI, and ResAI, and Galaxy Weblinks, each addressing a distinct class of real-world challenges, from enterprise automation to generative AI for productivity. He also plays a strategic advisory role with the Government of Karnataka, contributing to the digital transformation of public services and citizen access. But beyond roles and milestones, what sets Nilesh apart is his ability to bring structure to ambiguity and translate emerging technologies into scalable, people-first solutions. He isn’t in the business of hype. He’s building what truly matters, whether that’s real-time property validation in Bengaluru or quantised AI agents in global banks. Moving beyond understanding AI, Nilesh knows how to make it work responsibly and effectively, for platforms, for people, and for the public good. He shares his journey with TradeFlock in this exclusive interview.
What’s been the defining theme or turning point in your 20+ year journey?
The defining theme has been my willingness to pivot and build. I started with C++, went deep into operating systems, moved to Java services at Microsoft, then scaled cloud systems, and later stepped into finance tech at Goldman Sachs. And now, I’m fully immersed in AI, building agentic platforms and human-centric tools. Every phase required me to unlearn and relearn. But that’s been my edge; I don’t fear complexity; I organise it.
Smart AI isn’t just about power, it must be purposeful and pragmatic to deliver real value.
From your time at Microsoft, Amazon, and Goldman Sachs, what’s one principle you still apply?
My time at Microsoft, Amazon, and Goldman Sachs shaped my thinking in distinct ways. Microsoft taught me the value of system design through consensus—there’s real beauty in collective problem-solving. At Amazon, I learnt to innovate within constraints, finding ways to do more with less. Goldman Sachs instilled in me the discipline to build technology that can withstand financial and reputational risks at scale. Today, I carry a blend of these lessons: pragmatic thinking, clarity under pressure, and an unwavering focus on user trust.
How do you make high-stakes decisions—and reset after intense phases?
I always separate the decision from the ego. Whether it’s code I wrote or a direction I proposed, I stay open to critique. When the stakes are high, I run the scenario by trusted peers—then make the call and detach from the outcome. That detachment is essential to staying effective long-term. As for recovery—I ride. Long, unstructured solo rides through the Western Ghats or the Himalayas. No itinerary, no deadlines. It’s where I recalibrate.
What does success look like for a modern CTO?
It’s not just product metrics or ARR. Success means building capacity – within systems and within people. A great CTO sets a strategy, yes—but also nurtures a culture of curiosity, feedback, and thoughtful risk-taking. Technology will evolve. Your job is to ensure your people evolve faster.
What’s a key challenge tech leaders face today, and how can it become an opportunity?
There’s a growing illusion of expertise. AI is generating answers faster than people are learning the fundamentals behind them. Many leaders now optimise it for delivery velocity rather than depth of understanding. The danger is that if foundational knowledge erodes, innovation collapses. But here’s the opportunity: leaders who champion deep learning, of both the tech and the people, will be the ones to build enduring systems. That’s the philosophy behind ResAI, where we use LLMs not to replace learning but to personalise and accelerate it.
How do you ensure data privacy and compliance don’t clash with speed of product delivery?
I use a mental model borrowed from Amazon: “two-way door” vs “one-way door” decisions. Launching a beta feature? That’s a two-way door—you can iterate or roll back. Shipping with questionable GDPR compliance? That’s a one-way door. You break trust, and you may never recover. For me, if something involves a oneway door like security or privacy, I don’t I compromise—even if it delays the launch. It’s easier to recover from lost time than lost credibility.
What’s your current focus across JupiterBrains, CanbizAI, and ResAI?
At JupiterBrains, we’re building an agentic AI layer for enterprises. These are not just chatbots—they’re decision-support agents fine-tuned to legal work flows, investment research, fraud analysis, and more. CanbizAI is empowering India’s entrepreneurs. We’ve helped hundreds of founders go from idea to pitch deck to MVP in weeks—not months. ResAI is democratising access to upskilling. If you’re a student in Indore or an engineer in Hyderabad, ResAI helps you map your skill gaps and build a plan without depending on expensive coaching. All three platforms are bound by a common thread: purpose-built, efficient, and humanaligned AI.
Featured Magazine -
