Lessons Learned From Building and Scaling DietGhar and NyayX

Building two companies from zero to product-market fit taught me more than 14 years of consulting ever did. Here are the lessons that changed how I engineer.

Lesson 1: Boring tech wins

For DietGhar, I wanted to use the shiniest stack. I resisted. Postgres, Next.js, and a boring monolith got us to launch in 3 months. The startup graveyard is full of companies that chose microservices before they had 10 users.

Lesson 2: Observability from day one

With NyayX, I instrumented everything before writing a single feature. Error rates, p95 latency, conversion funnels — all visible on day one. This paid dividends immediately: I found a 30% drop-off in onboarding that I would have missed for months otherwise.

Lesson 3: Your infrastructure is a product decision

Every infrastructure choice has a user-facing consequence. Slow cold starts = users leaving. Expensive queries = slower features. I now review infrastructure costs monthly with the same discipline as revenue.

Lesson 4: Automation is leverage

Every manual deployment is a tax on your time. CI/CD, automated testing, infrastructure-as-code — these aren't "nice to haves." They're the difference between shipping once a week and shipping 10 times a day.

Lesson 5: Find your unfair advantage

Mine is engineering. I can build in weeks what agencies charge months for. Figure out what yours is, and stop leaving it on the table.