Infrastructure Engineer who builds systems that survive scale and failure.

Distributed Systems Platform Infrastructure AI-Native Tooling AI Research

At Microsoft I ran a distributed validation platform across 17,000+ microservices — reducing manual audit toil by 70%. At Amazon I built the device infrastructure running Alexa 24/7 across 50+ physical devices.

This still feels like day one. Harder problems. Larger scale. End-to-end.

// About

I'm an infrastructure engineer with 7+ years of end-to-end ownership — design through reliability. At Microsoft and Amazon I built platforms that thousands of engineers depend on without thinking about it. Self-healing systems. Deterministic automation. Infrastructure that disappears into the background.

Since late 2023 I've been conducting independent AI infrastructure research — shipping three open-source systems (Clairvoyant, ACO, ServiceScope) and preparing an arXiv preprint targeting MLSys 2027. Both are fully local, privacy-first, and built for production reality — not demo conditions.

I'm looking for Senior Infrastructure or ML Systems roles where I can own hard problems end-to-end. Based in Hyderabad — open to Bengaluru, Chennai, and international relocation (London, Singapore, Amsterdam).

Distributed Systems Event-Driven Architecture Concurrency Control Traffic Shaping Observability CI/CD Platforms Kubernetes Docker Terraform Device Orchestration Platform Reliability LLM Inference Infrastructure

// Work Experience

// Projects

// Active Research

// Writing

// Let's talk

Looking for Senior Infra roles where I can own hard problems end-to-end.

Open to distributed systems, platform infrastructure, and AI-native tooling roles. Based in Hyderabad — open to Bengaluru, Chennai, and international relocation (London, Singapore, Amsterdam). I write code, design systems, and care deeply about reliability.

aravindsundaresan099@gmail.com linkedin.com/in/aravind-sundaresan github.com/Aravind0403 aravindsundaresan.substack.com