Utsav Singh | Software Development Engineer 2
WORK EXPERIENCE
Flipkart | Software Development Engineer 2
March 2023 - Present
- Designed and developed the orchestration engine powering SuperMoney's co-branded card onboarding journeys (Axis, Kotak, Utkarsh). Achieved p99 latency of ~230 ms for onboarding and ~1.2 s for orchestration, outperforming competitors by >10x, enabling seamless integration with external bank APIs.
- Architected and solely built the lead-journey system that matches users with optimal lending partners using the orchestration engine handling ~10K+ journeys/day and facilitating issuance of ~2K credit cards/day.
- Created a cross-service Data Platform SDK to standardize event publishing to GCP Pub/Sub, adopted by 7+ teams, processing 250K+ events/day. Abstracted infra details behind a single pushEvent() API, cutting integration effort by >80%.
- Designed multi-card user support, requiring both vertical and horizontal redesign across onboarding and allocation systems.
- Improved oncall stability post Axis-SuperMoney launch: reduced noisy alerts by >95% (~1 every 15 min to ~1 every 2 hr), decreased journey failures by ~96%(from ~8% to ~0.3%), and resolved 50+ production issues.
- Developed marketplace gateway APIs for Flipkart Health+, ensuring seamless partner integrations.
- Improved inventory refresh latency from ~45 min to ~5 s using CDC pipelines and Redis, cutting out-of-stock incidents by ~50%.
- Implemented CDC pipelines with GCP Pub/Sub, Debezium, and gRPC, handling ~1.7 M messages/day.
- Established canary deployments with Helm & Kubernetes to enhance rollout reliability and reduced risk.
- Upgraded services from JDK 8 to 17, improving latency and build time; implemented Locust-based performance tests to validate NFRs.
Amazon | Software Development Engineer
December 2021 - March 2023
- Designed an auto-scalable serverless backend to delete ad keywords and related entities at ~864K deletes/day (~306 M/year) using AWS ECS, Fargate, S3, and SQS.
- Re-engineered the ad keyword sourcing algorithm, increasing effective keywords by ~185%, ad impressions by ~168%, and clicks by ~132%, driving ~$94 M (~48%) ad revenue growth.
- Enhanced ad-spend anomaly detection to cut false positives by ~57%, improving advertiser trust.
- Built 8+ chaos experiments to strengthen resiliency via improved resource configs, monitors, and alarms.
- Added support for modern ad assets (enhanced sitelinks, countdowns) and revamped stakeholder UI.
- Led Prime Day readiness activities, auditing and remediating compliance, operations, and security issues.
Ultragenic Research & Technologies | Software Developer
January 2021 - December 2021
- Built a serverless backend to extract data from CIOMS PDF forms, convert to E2B(XML), and feed into Oracle Argus Safety System, increasing throughput by ~80% and eliminating manual entry.
- Developed automated email processing service to parse E2B(XML) files, create safety cases, and generate reports, saving ~20 min per case (~160K cases/month).
- Conducted performance/load testing with Apache JMeter for client Argus web instances (HTTP/HTTPS).
- Stack: TypeScript, Python, PL/SQL.
LEADERSHIP EXPERIENCE
Enigma Programming Club | Head
April 2017 - July 2021
- Led a team of 80+ developers, organizing coding contests and workshops on DSA and C++.
- Developed the club's website to showcase achievements and drive engagement in the college tech community.
EDUCATION
ABES Engineering College
B.Tech in Computer Science & Engineering
Graduation Date: July 2021
- Core Courses: Data Structures, Algorithms, DBMS, OS, Networking, OOP, Functional Programming
SKILLS
- Programming Languages: Java, Python, C/C++, JavaScript, TypeScript
- Frameworks & Backend: Spring Boot, Dropwizard, Node.js, JUnit, Mockito, Microservices, Serverless
- Databases: MySQL, PostgreSQL, Redis, DynamoDB
- Cloud: AWS (ECS, Fargate, S3, SQS), GCP (Pub/Sub, GKE, BigQuery)
- DevOps: Docker, Kubernetes, Terraform, Helm, CI/CD Pipelines
- Tools & Collaboration: Git, GitHub, Jira, Confluence, Agile / Scrum
- Other: Data Structures & Algorithms, System Design, Distributed Systems, Observability, REST/gRPC APIs