Sudheer Bulusu

Backend Engineer • Distributed Systems

I am a backend engineer with 4 years of experience building distributed systems, workflow orchestration platforms and backend infrastructure for SaaS and consumer products.

Previously, I worked at Thena and Unacademy, where I designed event-driven services, recommendation systems, automation platforms and backend systems serving millions of users.

I am currently pursuing a Master's in Computer Science at UMass Amherst, where I am deepening my understanding of distributed systems, machine learning and large-scale software systems.

Experience

  • Software Development Engineer II · Thena

    Jun 2024 - Nov 2025

    Built workflow orchestration, CRM, notification and customer support systems powering a modern B2B customer support platform. Focused on distributed workflows, event-driven architecture, and platform reliability.

  • Software Engineer · Unacademy

    Jun 2022 - Jun 2024

    Built backend systems across multiple high scale learning products, including recommendation engines, subscription and billing systems, experimentation infrastructure, and real-time competitive learning platforms.

Experience details →

Education

  • Master of Science in Computer Science

    Jan 2026 - May 2027 (Expected)

    University of Massachusetts Amherst

    Amherst, MA

    GPA 4.0/4.0 · Distributed & Operating Systems, Advanced Algorithms, Software Engineering, Machine Learning, Neural Networks

  • Bachelor of Technology in Mechanical Engineering

    July 2018 - May 2022

    Indian Institute of Technology Bhubaneswar

    Bhubaneswar, India

    GPA 8.01/10 · Data Structures & Algorithms, Robotics, Advanced Calculus, Linear Algebra & Complex Analysis, Transform Calculus, Partial Differential Equations

Skills

  • Languages

    TypeScript · Go · Python · C++

  • Backend

    Node.js · Nest.js · Django · gRPC · Temporal

  • Data

    PostgreSQL · MySQL · Redis · DynamoDB · Elasticsearch · Redshift

  • Cloud

    AWS (ECS, SQS, SNS, Lambda, DMS) · Docker · GitHub Actions

Selected Work

  • Physics Informed Neural Networks (PINNs) for Crash Dynamics

    Built Physics-Informed Neural Networks (PINNs) to model vehicle crash dynamics, reducing simulation turnaround time from hours to near-instant inference while maintaining acceptable error bounds against analytical and FEM baselines.

  • RoomMatch

    Built a roommate matching platform that ranks candidates using lifestyle compatibility scores and Redis-powered real-time retrieval, enabling fast personalized recommendations.

View all projects →