Ankshilp Software
hello@ankshilp.in

I migrate monoliths without breaking production.

Anyone can generate a service skeleton now. The hard part was never writing the code — it's knowing which seams are safe to cut, what order to move things in, and what to leave alone until it's actually ready. That's eleven years of backend judgment, not a prompt.

Email hello@ankshilp.in Usually replies within a day.
monolith auth payments notifications billing

one deploy, one point of failure → independent services, deployed and scaled on their own

Three kinds of work, all backend, all in Go

Monolith to microservicescore focus

The code is the easy part. The real work is figuring out which pieces of your system are safe to pull apart, sequencing the move so nothing breaks in production, and knowing what to leave untouched for now. Done incrementally, alongside your existing system — not a rewrite from scratch.

New backend systems

Building a backend from scratch in Go for a product that's past the prototype stage and needs to handle real traffic, real data, and real failure modes from day one.

Performance & reliability

Diagnosing slow endpoints, memory leaks, flaky deploys, and scaling bottlenecks in an existing Go, Java, or Python backend, without a rewrite.

Who's doing the work

11 years writing backend systems in production
Go primary language for the last several years, plus Java and Python
1 person — you talk to the engineer doing the work, not an account manager

No team of juniors, no handoffs. If you hire Ankshilp Software, I'm the one reading your code and writing yours. Currently also learning Rust for systems-level work where Go's garbage collector gets in the way.

Something I built and run myself

Live product · built solo

webshield.in

A DNS filtering service built on DNS-over-HTTPS and DNS-over-TLS — it sits between a user's device and the internet, encrypting and filtering DNS lookups in real time to block ads, trackers, and malicious domains before a connection is even made.

It's the kind of system that has to be fast on every single request, has no tolerance for downtime, and deals directly with the networking and protocol-level detail that most backend work never touches.

DNS-over-HTTPS DNS-over-TLS Go low-latency infra

Three steps before any code changes

A conversation about the actual problem

A call or email thread about what's slow, what's fragile, or what you're trying to build — no proposal, no pitch deck, just figuring out if this is a fit.

A scoped assessment

I look at your current system (or requirements, if it's new) and come back with a specific plan: what would move first, what stays put for now, and a realistic estimate.

Incremental delivery

Work happens in shippable pieces, alongside your existing system where possible, so you see progress and can change course early rather than at the end.

Tell me what's slow, fragile, or missing

Send a few lines about your backend — what it does, what language it's in, what's not working — and I'll reply with honest thoughts on whether a migration or rebuild actually makes sense for you.