I Fired My CTO and Rebuilt Our Stack in 90 Days

For eight months I had been ignoring a feeling that kept waking me up at 2 AM. Every time a customer complained about a slow page, every time we could not ship a feature because of something buried in old code, every time I sat in a technical meeting and felt like I was being deliberately confused rather than genuinely explained to — the feeling was there.

My CTO and I had built this company together. We had been friends for six years. Letting him go was the hardest professional decision of my life.

But it was also the right one.

The Accumulation of Small Lies

Technical debt is not one dramatic moment. It is a hundred small decisions, each of which seems reasonable at the time. A quick fix here. A workaround there. We will refactor it later. Later never came.

By the time I understood the true state of our codebase — really understood it, by hiring an outside auditor and sitting with them for three days — we were sitting on three years of accumulated decisions that had made us fast to ship in year one and nearly impossible to scale in year three.

Our checkout was a Frankenstein of three different payment integrations, none of which talked to the others cleanly. Our inventory sync ran on a cron job that sometimes did not run. Our mobile site was a responsive desktop site that happened to work on phones, barely. We had seventeen third-party plugins that collectively cost us ₹3.2 lakhs a year and created seventeen different points of failure.

The Decision

I will not sensationalize this. Letting go of my CTO was painful for both of us, handled with as much dignity and care as I could manage. We are still, cautiously, friends. But the moment I made the decision, something shifted in me. I stopped waiting for someone else to fix the problem. I became responsible for it.

I gave myself 90 days. Not because 90 days is a magic number but because I had a board review in 90 days and I needed to walk in with a working platform, not a plan.

The Bet

The most important decision I made in those 90 days was not a hiring decision. It was a platform decision. I chose not to rebuild everything from scratch. I chose to move to a purpose-built commerce platform and let their engineering team carry the infrastructure weight that had been crushing us.

I remember calling QuantumOS X3's team on a Saturday. I explained our situation honestly: we were mid-rebuild, we had a deadline, we had a team of three engineers and a product manager, and we needed the platform to do heavy lifting so we could focus on what made us different.

They did not try to sell me. They asked me questions. What did our customers actually need? Where did our competitive advantage sit? What were we building on top of the platform, not just with it?

That conversation changed how I thought about the rebuild entirely.

What 90 Days Actually Looks Like

It is not glamorous. It is a lot of spreadsheets and Notion docs and late-night voice notes to yourself. It is prioritizing ruthlessly because you cannot do everything. It is making calls that turn out to be wrong and correcting them quickly instead of defending them.

We migrated our product catalog first — 4,300 SKUs, seventeen years of data, customer purchase history going back to 2008. Then the storefront. Then the checkout. Then, finally, the back office: orders, inventory, returns.

Day 87 we ran our first full end-to-end test. An order placed on the storefront, processed through the platform, picked from the warehouse, fulfilled, with a GST invoice generated automatically. It worked. First try.

I went home and slept for eleven hours.

The Board Review

I walked into that review with a live platform, a latency dashboard showing sub-100ms storefronts, and a new engineering budget that was 40% lower than the one we had been burning. The board asked me what had changed.

I told them the truth: I had stopped trying to build the infrastructure and started building the business.

That is the only real lesson here. Know the difference between what you should own and what you should trust someone else to have solved already. Ego is expensive. Clarity is compounding.

Subscribe to the QuantumOS Dispatch — weekly insights for commerce operators who want to compound their advantages.