Entrepreneurship
The Founder’s Calibration Practice
AI floods founders with confident answers and you quietly inherit its overconfidence. The founder's calibration practice trains your judgment to match reality.
Reinvent yourself for the AI age. I'll show you how I did it.
Reskill. Rewire. Reinvent.
I'm Vikas Malpani. I built CommonFloor and sold it for $200M, backed Unacademy before anyone else, then walked away and rebuilt myself AI-native. I've started over four times, on purpose. Everything I learn about reinventing, the skills, the mindset, the motivation to start, I write here. All of it free.
The best companies aren't built by chasing markets. They're built by people who see a problem so clearly, they can't not solve it. Everything else follows from obsession.
On building with conviction
Where do you want to start?
Not sure where you stand? Take the 2-minute Reinvention Readiness check →
Rewire how you think
The hard part of starting over isn't the skill, it's the identity. Rebuilding judgment, energy, and the will to move before you feel ready.
See where AI opens doors
Not which tools to learn. Where AI makes you more valuable, what it makes worthless, and where you can build something that matters now.
Build something of your own
From a sewage-pipe complaint to a $200M exit, and everything I got wrong along the way. The playbooks for starting, scaling, and not quitting.
From a Sewage Problem to a Real Estate Revolution
In 2007, a sewage problem in my Bangalore apartment building led me to discover something unexpected: my school friend Sumit Jain lived next door. We spotted the real problem: India’s broken real estate market had no centralized information.
One year after graduating from VTU, Sumit and I founded CommonFloor to fix this. What started as a simple idea became India’s largest real estate marketplace — with millions of listings, 200+ cities, and eventually a $200M exit.
Read the full storyEssays
Field notes from four reinventions — lessons from a $200M exit, two unicorns, and building AI-native, published as I learn.
Start Here
Entrepreneurship
The Solo Founder AI-Native Operating System
The Solo Founder AI Operating System is the six-loop model that lets a one-person company survive vendor price shocks and still ship AI that actually works.
Building
Why I Left a $200M Exit to Start Over
The hardest part of entrepreneurship is not building — it is walking away from something that works to chase something that matters more.
AI & Technology
The Inference Shift: Where AI’s Real Cost Moved
The inference shift moved AI's real cost from training models to running them. Where the money, the constraint, and the founder opportunity actually live now.
Entrepreneurship
The Absorption Test: Will AI Absorb Your Business?
The real moat question is not whether AI can do what you do. It is whether anyone still pays you after it does. Here is the Absorption Test and how to pass it.
Entrepreneurship
The Initiative Problem: The Skill AI Quietly Erodes
AI is great at the how and silent on the what. The skill that atrophies is initiative: deciding what to do when nothing prompts you. Here is how to protect it.
Entrepreneurship
AI Agent Costs: When Headcount Becomes a Variable Bill
AI agent costs do not just shrink payroll. They change its shape, turning a fixed salary into variable metered compute. Here is the margin trap and the fix.
AI & Technology
Why AI Agents Ship in Regulated Work First
AI agents reach production in regulated, auditable work first. The moat is the audit trail, not the model. Here is the deployment order and why.
Entrepreneurship
AI Burnout: Why the Bar for Enough Keeps Rising
AI burnout is not about working too little or resting too much. It is the bar for enough rising faster than you can meet it. Here is how to define enough.
Entrepreneurship
Revenue Per Employee: The Signal That Inverted
Revenue per employee became the number investors trust, and headcount growth turned into a warning sign. How the signal inverted, and how to hire lean.
Track Record
Three companies. Two unicorns. One pattern: build technology that moves markets.
$200M Exit 2007 – 2016
maxHeap / CommonFloor
Co-Founder (with Sumit Jain & Lalit Mangal)
Co-founded maxHeap with Sumit Jain and Lalit Mangal. Launched CommonFloor.com and transformed India’s real estate market. $69.6M raised across 7 rounds from Tiger Global, Accel Partners, Google Capital, and others. Merged with Quikr in a $200M deal.
Active 2023 – Present
Garvik AI / ReBillion.ai
Co-Founder & CEO (with Atul Jaju)
Building ReBillion.ai — enterprise AI middleware enabling companies to deploy company-wide knowledge search and AI copilot capabilities. AI assistants deployed across Banking, Manufacturing, Real Estate, and EdTech verticals.
Exited 2018 – 2022
Leher
Co-Founder & CEO (with Atul Jaju, ex-Goldman Sachs)
Pioneered live audio-video social networking in India. Backed by Google for Startups Accelerator (2020) and Orios Venture Partners.
Investing
Early-stage bets on exceptional founders building at India-scale. Two unicorns and counting.
Unicorn
Unacademy
Earliest Angel Investor
Seeded Unacademy in December 2015 with a $586K round. The company has grown into India’s most valuable ed-tech unicorn, now valued at $3.4B+.
Unicorn
Quikr
Via CommonFloor Merger
CommonFloor’s $200M all-stock acquisition by Quikr in 2016 created India’s leading classifieds platform, now a unicorn.
Investment Thesis
Founder-First
I back exceptional people, not just pitch decks. Founders who are obsessed with the problem, not the fundraise.
India-Scale Problems
Technology solving problems unique to India’s scale: real estate, education, enterprise AI, and emerging sectors.
Early & Active
Seed and pre-seed. I don’t just write checks. I roll up my sleeves: strategy, hiring, product, GTM.
Advisory & Speaking
Strategic advisor to high-growth startups. Operator experience, not just opinions.
Product & Strategy
What to build, what to kill, and how to find product-market fit in complex markets. Done this four times.
Scaling & GTM
From zero to 200+ cities. I know what it takes to scale across India: distribution, partnerships, and growth loops.
Fundraising & Exits
Raised from Tiger Global. Executed a $200M exit. I help founders navigate from seed to Series B and beyond.
The method: the Reinvention OS
I've started over four times, on purpose, and out of it I built one repeatable system for reinventing yourself when the ground moves under you. I call it the Reinvention OS, and I'm writing the whole thing down here, free. Three moves, in order:
1. Mind before method. Rebuild judgment, energy, and the will to move before the skills. The identity is the hard part, not the tools.
2. Aim at the openings. Find where AI makes you more valuable, not less. Most people learn tools. Reinventors find leverage.
3. Ship to become. You reinvent by building, not by planning. Small shipped things compound into a new you faster than any course.
Everything here is free. Here is the deal.
No ads. No selling your data. No upsell sequence waiting in your inbox. I make my living building companies, not building a list, so I have nothing to sell you. The only thing I ask: if something here moves you, build something, and tell one person where you got the push.
Join the Reinventors and get each essay as I write it.
Reinvention, answered
How do I reinvent my career for the AI age?
Start with mindset, not tools. Rebuild your judgment and energy, find where AI makes you more valuable, and ship small projects that compound into new skills. That sequence is the Reinvention OS.
Is it too late to start over at 40 or 50?
No. Mid-career is an advantage, not a handicap. You carry judgment, relationships, and pattern recognition a 22-year-old does not, and the AI age rewards exactly that.
Do I need to learn to code to be AI-native?
No. Being AI-native is an operating habit, not a coding skill. It means using AI to think, decide, and build faster. Judgment is the bottleneck, not syntax.
Is everything on this site really free?
Yes. Every essay and framework is free, with no paywall and no upsell. I make my living building companies, so the writing is a gift, not a funnel.