AI Product Manager MBA Roadmap: How Sunil Cracked Google’s ₹35LPA Offer
From confused engineer to Google AI PM — one man’s honest, unglamorous, and deeply intentional journey.
Every second week, someone posts a LinkedIn victory lap. A shiny offer letter. A Google or Microsoft logo. A number with “LPA” at the end. And then comes the flood of “How did you do it?” comments that never quite get a real answer.
This is Sunil’s real answer.
Sunil Mehta, a mechanical engineer from Nagpur with four years of experience in an automotive firm, landed a ₹35 LPA offer as an AI Product Manager at Google India. No IIT pedigree. No prior tech background. No referral from a college senior who “just happened” to be at the company. What he had was a plan — built slowly, tested repeatedly, and executed with uncommon patience.
Here is the roadmap that actually worked.
Starting Point: The Problem With Most PM Aspirants
Before we get into what Sunil did right, it is worth understanding what most people do wrong.
The typical MBA aspirant who wants to break into product management reads one book (usually Inspired by Marty Cagan), takes a weekend crash course, memorizes a few frameworks like CIRCLES and RICE, and then walks into interviews expecting to compete with engineers who have shipped real products for years.
Sunil knew this was a losing game. His first decision — and perhaps his most important one — was to slow down when everyone around him was rushing.
“I gave myself 18 months,” he said. “Not to get the job. To become someone who deserved the job.”
That mindset shift changed everything.
Phase 1: Building Technical Credibility (Months 1–5)
Sunil’s engineering background gave him an advantage he almost ignored. Most non-tech MBA candidates spend months trying to prove they can “speak tech.” Sunil already could — he just needed to redirect that fluency toward AI and software systems.
He began with a structured self-study plan:
- Andrew Ng’s Machine Learning Specialization on Coursera (not to become a data scientist, but to understand how models are trained, evaluated, and deployed)
- SQL and basic Python — enough to pull data, write simple scripts, and not embarrass himself in a technical conversation
- System design fundamentals — understanding APIs, microservices, latency trade-offs, and what engineers mean when they talk about “scale”
He spent roughly three hours every evening on this, alongside his day job. No shortcuts. By month four, he could read an ML model card, understand feature importance, and ask intelligent questions in a product review meeting — not just nod along.
This phase was not glamorous. There were no certificates worth posting. But it laid the foundation that made everything else possible.
Phase 2: The MBA — Choosing Right, Not Prestige-Chasing (Month 3–8)
While building his technical base, Sunil simultaneously prepared for MBA admissions. His target was not the highest-ranked school he could get into. It was the school whose programme best prepared him for product roles in tech.
He eventually joined a Tier-1 MBA programme with a strong emphasis on technology management and had active recruiter relationships with Google, Amazon, and Flipkart. He chose it over a marginally higher-ranked school that had weaker tech placement data.
During the MBA itself, he made three strategic choices that most of his peers did not:
1. He chose his electives with surgical intent. While classmates loaded up on finance and marketing courses, Sunil took every technology strategy, data analytics, and AI ethics elective available. He understood that in product interviews, what you know specifically matters far more than what you know broadly.
2. He built in public. He started a modest newsletter — not for followers, but for accountability. Every two weeks, he wrote about an AI product he had dissected: how it worked, what problem it solved, where it fell short, and what he would do differently. This became his portfolio.
3. He got his hands dirty. Through the MBA’s entrepreneurship cell, he co-managed a student-run app project. He wrote user stories, ran sprint planning sessions, talked to actual users, and made the call to kill a feature that users loved but that killed retention. That single experience gave him more interview material than six months of case prep.
Phase 3: Interview Preparation — The Method Behind the Clarity
Google’s PM interviews are not gentle. They test product sense, analytical thinking, leadership, and technical depth — often in the same question.
Sunil’s preparation had three pillars:
Structured practice, not random prep. He worked through 150+ product questions over four months, always writing his answers before speaking them. Writing forced clarity. He used a modified STAR format for behavioural questions and a custom framework for product design questions that started with user empathy before jumping to solutions.
Mock interviews with people who would be honest. Not friends. Not family. He found two senior PMs through LinkedIn who agreed to give him brutal feedback. One of them told him after his first mock session that he “sounded like a consultant, not a builder.” That stung — and it was exactly the correction he needed.
Google-specific intelligence. He studied Google’s publicly stated product principles, read every available interview debrief on Glassdoor and Blind, and mapped his experiences to the specific things Google’s interviewers care about: user empathy, data-informed decisions, and cross-functional leadership.
The Offer — And What It Actually Represents
When the ₹35 LPA offer came through, Sunil did not post immediately. He sat with it for a day.
“I wanted to remember that the number was a side effect,” he said. “The real thing was that I had become someone who could genuinely do the job.”
That is the part of the story that never makes it into the LinkedIn post. The 18 months of deliberate, unglamorous work. The evenings spent on Coursera when friends were unwinding. The newsletter that had 60 subscribers. The mock interview feedback that made him want to quit.
The Roadmap, Distilled
For anyone charting a similar path, Sunil’s journey points to five clear principles:
- Technical credibility is non-negotiable. You do not need to code, but you must understand what your engineers are building and why it is hard.
- Choose your MBA programme for placement fit, not ranking alone. Research where each school’s tech PM alumni actually land.
- Build a portfolio of thinking, not just credentials. Case studies, product teardowns, and side projects tell interviewers more than your CGPA.
- Get honest feedback early and often. Comfortable practice produces comfortable failure in real interviews.
- Play a long game. Eighteen months of consistent effort beats six months of intense cramming — every time.
Sunil’s story is not exceptional because he is exceptional. It is exceptional because he was patient when the system rewards speed, specific when everyone else was broad, and honest with himself when self-delusion would have been easier.
That, more than any framework or course, is what cracked Google.