Suraj C Jagatap

About

For seven years I've built and looked after infrastructure on AWS — provisioning it, automating it, watching it, and fixing it when something breaks at three in the morning.

Most of that work has been in distribution and retail, where a few hours of downtime means trucks don't move and orders don't ship. That taught me to care about recovery more than elegance. A system that fails quietly and comes back fast beats a clever one that surprises you.

I'm self-taught. Everything I know came from building things, breaking them, and being on call for the result. These days my time goes on multi-account AWS setups, Kubernetes, Terraform, and cutting cloud bills that grew while nobody was watching.

Experience

• Now

  Senior Software Engineer, Site Reliability

  Infinite Locus

  Running infrastructure across several client accounts — provisioning, monitoring, cost work, and the day-to-day of keeping other people's production systems healthy.

  AWSKubernetesTerraform
• Previously

  Infrastructure & Hosting

  Intelligent Retail (Ripplr)

  Owned hosting and infrastructure for an FMCG distribution platform — the systems that move stock and orders for a physical supply chain.

  AWSDockerLinux
• Earlier

  Operations & Logistics Systems

  Aditya Birla

  Inventory and inward/outward flows, internal file transfer, and keeping cloud APIs healthy. Where I learned what downtime costs a business that ships real things.

  OperationsAPIs

Things I built

• Internal

  Multi-account health monitor

  One dashboard showing the health of every AWS account we manage, instead of logging into each one to find out something had already broken.

  PythonCloudWatchboto3
• Mobile

  PhysioBook

  A scheduling and records app for a home-visit physiotherapy practice. Built end to end, from the data model to the store build.

  FlutterDart
• Personal

  Agent workflow system

  A two-layer system for working with AI coding agents that doesn't lock me into one tool. Runbooks for machines, essentially.

  PythonLLM tooling

Stack

Now

AI engineering, pointed at operations.

Retrieval over runbooks, agents that can execute them, and the reliability practice that has to sit around a model once it's in production. It maps cleanly onto work I already do — a model endpoint is an API with worse failure modes, and LLMOps is mostly SRE wearing a new hat.

Alongside it I'm working toward the AWS DevOps Engineer Professional, CKA, and Terraform Associate.

Contact

Infrastructure that needs building, watching, or rescuing?

Write to me at hello@mgsr.dev, or find me on LinkedIn.