Nrishinghananda Roy.

A performance problem that needed a real fix, not a workaround

At Codefy GmbH in Germany, a third-party Java microservice was adding over 3 seconds of latency to every API request. I read through the open-source codebase, understood exactly what it was doing, and rebuilt it from scratch in Rust (Axum) and PostgreSQL. API response times dropped to under 200 milliseconds under production load. That is a 15x improvement on a live system.

A zero-downtime cloud migration, 60+ nodes, two platforms

At Dhiway, I provisioned and operated a globally distributed network of over 60 nodes across AWS EC2 and GCP using Terraform. When the decision came to migrate fully to GCP, I executed the migration while production stayed live. Centralized observability with Prometheus and Loki ran across the entire network throughout. No incidents. No service disruption.

Linux infrastructure and operations from day one

At Trustforum in France, I managed Linux server environments for an early-stage platform: user permissions, package management, system updates, log rotation, and automated backups via Bash scripting. The kind of unglamorous work that keeps systems healthy and teams uninterrupted.

Most infrastructure problems are not technical problems. They are judgment problems. Knowing which tool to reach for. Knowing when a system is healthy versus quietly failing. Knowing that the migration plan that looks clean on paper will not survive contact with production unless you have tested every assumption.

I have operated infrastructure in conditions where things going wrong had real consequences. That shapes how I approach the work. I do not build fragile systems and hope nobody notices.

I also think about security at the infrastructure level, not as an afterthought. Earlier this year I reverse-engineered a multi-stage malware attack targeting developers through a fake job interview. Tracing that attack chain changed how I think about container hardening, secrets management, and what a compromised developer environment actually looks like from the inside. That kind of thinking is built into how I design systems now.

Rust Clippy

Clippy is the official Rust linter used across the entire Rust ecosystem. I contributed lints and improvements to the project and am resuming active contribution now.

I am open to remote roles in platform engineering, DevOps, infrastructure engineering, and SRE. I am also available for serious freelance infrastructure engagements.

I have worked remotely for companies in Germany, France, and India. I know how to operate without hand-holding, communicate clearly across time zones, and deliver work that holds up.