Resume

ℹī¸

Urjit Singh Bhatia

STAFF SOFTWARE ENGINEER

Software Engineer with 15+ years of expertise in high-performance computing, AI, & cloud platforms. Led cross-functional initiatives, mentored teams, and developed both technical and product strategies. Proficient in high-throughput scaling, backend & distributed systems, security, & legal compliance. Architected systems supporting 30M+ concurrent users. Languages: English, Punjabi, Hindi, Gujarati, Urdu & beginner Norwegian.

San Francisco, CA urjitsb87@gmail.com | (703) 608-9877
LinkedIn: linkedin.com/in/urjitsinghbhatia | GitHub: github.com/urjitbhatia

Employment

ClassDojo Inc - May 2015 to Current

Staff Software Engineer - San Francisco, CA

  • Led AI initiatives, developing education-focused AI products.
  • Built a streaming AI pipeline for fake account detection and content moderation.
  • Spearheaded backend, security, infrastructure, performance tuning, and budgeting initiatives.
  • Architected a seamless payment system upgrade with zero user impact and downtime.
  • Designed a high-performance, image resizing service, reducing server instances from 80 to 6, significantly reducing large instance cloud costs.
  • Eliminated 53ms DNS resolution delay-per-request with a custom Node.js cache, further optimized with CoreDNS.
  • Developed a notifications infrastructure handling 500 million daily events across SMS, email, and push.
  • Ensured GDPR and regulatory compliance, collaborating with third-party vendors and legal teams and onboarded the CISO and their team.
  • Enhanced CI/testing infrastructure, achieving 4x speed improvements for backend and Android services.
  • Established data analytics functions, collaborating on product strategy with PMs.
  • Mentored teams, led scaling initiatives, and developed distributed systems aligned with business needs.
  • Engaged in user research and represented the organization at industry conferences as a speaker and exhibitor.

Knewton - Mar. 2014 to May 2015

Senior Software Engineer - New York, NY

  • Led development of Graph ingestion service handling concurrent graph operations and complex ontological validations.
  • Implemented a novel 3-way graph data merge algorithm.
  • Developed services which handled partitioned graphs containing 60-100k vertices, shared taxonomies clocking at 10’s of GB of data distributed over multiple Cassandra servers involving optimistic node hops.
  • Designed and developed a plugin-based load testing system capable of generating novel load profiles and replaying terabytes of production data.

Knewton - May 2012 to Mar. 2014

Software Engineer - New York, NY

  • Member of team that developed service platform to handle nearly 700 transactions per second from partner’s event streams.
  • Developed automation toolchain for handling AWS stack deployment using CloudFormation & Boto.
  • Scaled service platform by identifying performance bottlenecks and tuning caches.

Next Jump - Feb. 2011 to May 2012

Software Engineer, Data Tracking and Transactions

  • Developed platform services for partner integrations, trend forecasting & financial data processing.
  • Responsible for technical communication with partners for business integration and API development.
  • Designed and implemented credit card fraud detection protocol, with experience in PayPal, Mastercard, Amex, and Visa payment gateways.

Next Jump - June 2010 to Aug. 2010

Software Engineering Intern

  • Developed a custom Product Search stack using Apache Solr + Lucene, handling upwards of 85-million items for search indexing.
  • Won best intern project award.

TCS - Jan 2009 to July 2009

Software Engineer

  • Developed an hospital management system.
  • Upgraded system to handle multiple uploads simultaneously saving Doctors precious report filing time, improved page load performance.

Education

Georgia Institute of Technology

Master of Science in Computer Science, 2010

Focus: Machine Learning & Systems

Gujarat University

Bachelor of Engineering in Information Technology, 2009

Focus: Software Engineering, Systems, Graphics, AI/ML

Skills

Cloud Platforms: AWS, GCP

Backend Development: High Throughput Systems, Distributed Systems, ML Ops

Machine Learning & AI: Data Analytics, AI, Transformers, PyTorch, OpenCV, LLMs, Vector Databases

Languages: Go (Golang), Java, Python, Rust, Elixir/Erlang, C/C++, Bash

Datastores & Tools: MySQL, Postgres, Redis, MongoDB, RabbitMQ, Docker, Hashicorp Suite, Git

Projects

Cozgo

Read the blog post

cozgo 141 is a Golang wrapper for Coz - a Causal profiler based on the paper:
https://www.sigops.org/s/conferences/sosp/2015/current/2015-Monterey/printable/090-curtsinger.pdf .

Compared to traditional profiling highlights how to optimize hot-spots, Cozgo helps find what to optimize in a code base for the most ROI.

ChronoMQ

ChronoMQ 6 is a High-Throughput cancelable, delayed message queue inspired by the Beanstalkd protocol. Handles up to 30K RPS on commodity hardware with sub-millisecond ops latencies.

GoPipe

GoPipe 4 A stream-filter like pipeline primitive for Golang.

See more projects here…

Publications

CBArch: A Case-Based Reasoning Framework for Conceptual Design of Commercial Buildings

The paper describes the first phase of development of a Case-Base Reasoning (CBR) system to support early conceptual design of buildings with a focus on sustainable design
https://www.aaai.org/ocs/index.php/SSS/SSS11/paper/viewFile/2494/2934

Forensic Signature Verification

Comparison of various Machine Learning approaches towards automated signature verification based on the CEDAR signatures dataset
https://www.academia.edu/download/30816601/ForensicSignatureVerification.pdf

Volunteering

TogetherSF

Current
Volunteer with local residents to help clean up the city.

Surat Fauj Running Club

Member
Member of running group that supports various social causes and volunteers at local events.

Activities

Turtle Hacks - 2024

Judge
Volunteered as a judge for TurtleHacks - a 36-hour green-centric hybrid hackathon in Canada.

Computer Society of India

Student Lead - CITC Branch
Organized and initiated the “Special Interest Groups” amongst students to help them partake in going beyond the curriculum to learn by self motivated projects.

Microsoft

Student Partner India Region Lead
Organized and taught free of cost, peer-education workshops across the country. Delivered lectures, hands-on-lab sessions to provide technology access to engineering students. Specially focused on XNA game design and C# application development.