Nrishinghananda Roy.
Systems engineer working toward data infrastructure in Rust.
I have production experience in Rust backend services and distributed cloud infrastructure. My current focus is the internals of storage engines, query systems, and observability backends. I read production Rust codebases, work toward contributing to open-source projects in this space, and write about what I find.
What I am working on
I am studying the internals of data infrastructure and observability systems built in Rust, working toward open-source contributions in this domain. My method is self-directed: I read source code, trace data paths end to end through unfamiliar systems, and form architectural opinions before looking at how the project's authors resolved the same problems.
The systems I focus on sit at the intersection of storage engine design, query execution, and high-throughput observability: how log-structured storage engines manage compaction strategies to balance read and write amplification, how columnar formats exploit memory layout to accelerate analytical workloads, and how observability backends sustain high ingestion rates without query latency degrading under cardinality pressure.
Contributions to open-source Rust infrastructure projects in this space are in progress.
Selected work
Java to Rust service migration
At Codefy GmbH I read an open-source Java microservice and its PostgreSQL schema in full, then rebuilt the entire service in Rust using Axum and Tokio. The work required understanding an unfamiliar codebase from first principles, replicating its behavior correctly, and producing something measurably better. API response times dropped from over three seconds to under 200 milliseconds.
Document processing pipelines in Rust
At Codefy GmbH I built automated legal document workflows in Rust: PDF editing, image format conversion, page-level image placement, and DOCX generation from form data, using lopdf, image, and nom. The interesting parts required reading library source code directly because documentation stopped where the actual problems started.
Distributed infrastructure at scale
At Dhiway I provisioned and operated a globally distributed network of 60-plus nodes across AWS EC2 and GCP using Terraform. Set up Prometheus and Loki for observability across all regions, built GitHub Actions pipelines for deployment automation and security scanning, and executed a full cloud migration from AWS to GCP without significant service disruption.
Open source
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.
lsmdb
A persistent key-value storage engine built on the Log-Structured Merge Tree architecture in Rust. Implements a SkipList MemTable, Write-Ahead Log, Bloom filters, LRU block cache, and Snappy compression. Published on crates.io as both a library and CLI binary.
Writing
I write about Rust and storage system internals, and what I learn from reading production codebases. The goal is to make the design decisions inside complex systems legible to engineers approaching these projects for the first time.