Gopi Krishnan.

Where the work happened

Engineering principles I hold

• // on system design I design for failure before I design for success.▾

  The happy path is obvious — any engineer can make that work. It's the retry storms, the cascading failures, the edge cases at 2am that reveal whether an architecture actually holds. Complexity earns its place. It doesn't get assumed in.

• // on integrations Every integration will fail. The only question is whether your system recovers or corrupts.▾

  Retry logic and idempotency aren't advanced topics — they're the minimum bar. I treat failure handling as a first-class design requirement, not something you add after the first production incident teaches you why it matters.

• // on broken systems The first job isn't to fix it. It's to contain it.▾

  Assess blast radius. Stabilise. Communicate honestly — no speculation, just facts and timelines. Then find the root cause, and more importantly, find the observability gap that let it go undetected. Fix the system. Then document it — because the next engineer deserves a map, not a mystery.

• // on shipping Done means documented.▾

  I don't consider a project closed until the decisions, outcomes, and architecture are written down — whether it succeeded or failed. A clear diagram sent over email has saved more client relationships than any feature ever has. Clarity is a deliverable.

• // on complexity Simple solutions are harder to build than complex ones.▾

  Anyone can add abstraction layers, queues, and microservices. The discipline is knowing when not to. The best systems I've built look obvious in hindsight — which means all the hard thinking happened before the first line of code.

The things that have stayed true

• 01 Closure is non-negotiable.▾

  I don't consider anything done until there's a conclusion — written down, shared, and acted on. An unfinished thing isn't work in progress. It's just debt.

• 02 If you're going to do it, do it right.▾

  Seivana thirundha sei. Thoroughness isn't perfectionism — it's respect for the problem. A fast solution that half-works will cost you more than a slow one that works completely.

• 03 Plant perspectives. Don't give advice.▾

  Advice tells people what to do. A perspective gives them a new way to see — and lets them arrive at their own answer. The second one sticks. The first breeds dependency.

• 04 Filter ruthlessly. Absorb selectively.▾

  There's a Tamil saying about the Anna paravai — a bird that drinks only the milk from a mixture of milk and water, leaving the water behind. I try to consume information, feedback, and energy the same way. Take what builds you. Leave the rest.

• 05 Discipline first. Consistency follows.▾

  Motivation comes and goes. Discipline is a system you build once and run forever. Consistency isn't the goal — it's what happens when the system works.

• 06 Stay grounded. Stay confident.▾

  The world is incomprehensibly larger than any one person's knowledge. That's not a reason for self-doubt — it's a reason to stay curious, stay humble, and keep going anyway.