I am Nrishinghananda Roy, a Rust Systems Engineer focused on distributed infrastructure.
I write production Rust: async services with Tokio, gRPC with Tonic, HTTP APIs with Axum. I have built a persistent storage engine with a Write-Ahead Log in Rust, and operated Prometheus and Loki observability stacks across 60+ distributed nodes in production.
I am an active contributor to Rust Clippy, the official Rust compiler linting tool.
My work focuses on improving system performance, reducing operational overhead, and building reliable infrastructure that teams can depend on.
Built a persistent key-value storage engine from scratch in Rust, published on crates.io as a library and CLI binary. Implemented Write-Ahead Logging for crash recovery, Bloom Filters, LRU block caching, and an arena-allocated SkipList.
Provisioned and operated 60+ globally distributed nodes using Terraform and AWS EC2. Configured Prometheus and Loki to monitor node health and log aggregation across all regions. Built GitHub Actions pipelines for automated deployment and security scanning.
Replaced a Java microservice with Rust on Axum and Tokio, rebuilding APIs and cutting response times from 3+ seconds to under 200ms. Automated document generation workflows with custom PDF and DOCX parsing in Rust, reducing manual effort by 60% and accelerating processing 7x.
I write Rust at the systems level: async runtimes with Tokio, HTTP servers with Axum, gRPC services with Tonic, concurrent data structures, and zero-copy I/O. My focus is correctness, performance, and predictable behavior under load.
I understand how storage systems are built from first principles: Write-Ahead Logs, MemTables, SSTable compaction, Bloom Filters, and columnar formats. I have implemented these in Rust and read production storage systems, including OpenObserve's WAL and DataFusion-based query engine, as primary sources.
I study distributed systems from primary sources: Designing Data-Intensive Applications, the OpenObserve cluster codebase, and NATS-based coordination patterns. I understand replication, partitioning, consistency tradeoffs, and what happens when nodes fail, not just in theory but in the code that handles it.
I have operated Prometheus and Loki stacks in production and studied OpenTelemetry from protocol to implementation. I understand the full pipeline: SDK instrumentation, OTLP transport, collector configuration, and backend storage. I contribute to OpenObserve, an OTel-native observability platform written in Rust.
I understand the business context behind the systems I build. I take full ownership of features, making practical technical decisions that align with real product goals and fast-paced startup timelines.
I work remotely and autonomously. I have delivered production systems for teams in Germany, France, India, and the USA without on-site presence. I communicate technical decisions in writing, give and receive code review, and take ownership from design through deployment.
Freelance
Codefy GmbH | Remote, Germany
Dhiway | Remote, India
Trustforum | Remote, France
Engineered a persistent Key-Value storage engine from scratch in Rust. Replaced traditional database wrappers with a custom engine built strictly on zero-cost abstractions, multi-level compactions, and an arena-allocated SkipList.
Shipped to App Store in weeks with zero prior mobile experience. Handles festival info, real-time volunteer management, and role-based auth. 100+ downloads.