YMB Vadodara · IST

Yashodhan. Mohan Bhatnagar

engineer, currently leading engineering at Superr. Previously at Lyearn and Amazon.

A short introduction.

I'm an engineer who's spent the last decade building on the internet — equally at home with old-school Java (Spring, Tomcat) and the newer JavaScript/TypeScript stack (MEAN, React). I tend to gravitate toward problems that sit at the seam between systems: realtime collaboration, infrastructure that has to behave under stress, platforms that other teams build on top of.

B.Tech. (Honours) in Information & Communication Technology with a minor in Computational Sciences from DA-IICT. Research roots in high-performance computing and system modelling — those instincts still show up in how I think about software at scale.

Where I've been.

  1. 2024 — now

    Head of Engineering

    Superr · Mumbai

  2. 2020 — 2023

    Product Engineer IV

    Lyearn · Gandhinagar

  3. 2019 — 2020

    Software Development Engineer II

    Amazon · Bengaluru

  4. 2017 — 2019

    Software Development Engineer

    Amazon · Bengaluru

  5. 2017

    Data Science Intern · Teaching Assistant

    Playpower · DA-IICT · Gandhinagar

  6. 2016

    SDE Intern

    Amazon · Bengaluru

A few things worth pointing at.

  • Realtime collaborative editor

    Lyearn · backend lead

    Designed and led a Yjs-backed CRDT block editor with optimal, no-latency saves and tightly bounded DB writes. Built the block primitives as an extensible ecosystem so other product surfaces could build on top.

  • Low-bandwidth video conferencing

    Lyearn · platform

    Forked and customized Jitsi on a self-built Kubernetes cluster, tuned for the realities of Indian networks — encoding, delivery, and WebRTC paths reshaped to keep calls usable on thin pipes.

  • GCP landing zone in Terraform

    Lyearn · infrastructure

    Built a shared-VPC, air-gapped landing zone with east-west traffic through GCP and third-party firewalls. Networking, storage, containers, IAM — all codified, all reviewable.

  • Zero-downtime cluster split

    Lyearn · infrastructure

    Separated a shared dev/prod Kubernetes cluster into independent region-based clusters with no user-visible or developer-visible disruption. Planned the failure modes before the migration, not during.

  • Product Detail Page, at Amazon scale

    Amazon · core platform

    Core team behind one of the internet's heaviest pages. JVM and metaspace optimizations across 1000+ server fleets translated to multi-million-dollar savings; vended APIs and contracts to 100+ partner teams.

  • Let's HPC

    Research · 2018

    Co-author of a web platform for parallel and high-performance computing education, published in the Journal of Parallel and Distributed Computing.

Words.

See all writing →