Summary
Senior Backend and ML Engineer with 13+ years of experience designing and building
distributed systems across retail, banking, and enterprise software. Currently a Java Backend Engineer at
Walmart, where I design and build LLM/RAG-based backend systems that let users query large document sets
and get grounded answers — Hugging Face models for embeddings and LLM workflows, Qdrant for semantic
search, and ZenML to orchestrate evaluation pipelines that measure retrieval and generation quality.
My core stack is Java, Python, Apache Spark, and Apache Kafka on AWS, with a track record of
high-throughput, low-latency services — including a Kafka pipeline that increased throughput 10x.
I also design the backend system for inklink.ai using AWS and Java,
with AWS Lambda for infrequent services.
Experience
Java Backend Engineer — Walmart (Contract)
Apr 2025 – Present · Sunnyvale, CA
- Design and build LLM/RAG-based backend systems in production, enabling semantic search and grounded generation over large document sets.
- Built embedding and semantic-search workflows using Hugging Face models with Qdrant as the vector store.
- Built ZenML-orchestrated evaluation pipelines that measure retrieval and generation quality and catch quality regressions before release.
- Develop high-throughput, low-latency Java services on AWS around the pipeline, including Kafka-based event processing.
Backend System Design — inklink.ai
Ongoing
- Design the backend system using AWS and Java, with AWS Lambda for all infrequent services.
Senior Software Engineer — City University of Seattle
Jul 2024 – Apr 2025 · Seattle, WA
- Created a Java-based test automation framework for student reports, enhancing grading efficiency, with frontend development in JavaScript and comprehensive logging.
Senior Software Development Engineer — Visma Consulting
Aug 2021 – Sep 2022 · Helsinki, Finland
- Modernized a legacy application from a monolithic to a microservices architecture using Docker and Kubernetes, reducing development, maintenance, and server costs by 20%.
- Developed a new module from scratch using Ruby on Rails for digital signature customers, enabling new APIs to be created significantly faster.
- Automated build and deployment with a Jenkins pipeline, reducing lead time from 8 hours to 1 hour; implemented GitOps principles for CI/CD.
- Enhanced data security with a GDPR-compliant design addressing plaintext storage of user-sensitive data.
- Implemented version-agnostic backward and forward compatibility for REST services, eliminating client-side version adaptation issues.
Lead Software Engineer — Clairvoyant
Apr 2020 – Apr 2021 · Pune, India
- Implemented GDPR compliance for PayPal's large Hive data warehouse with a PySpark job masking specific primary keys — 2 million records processed in 30 minutes despite Hive 1.1 limitations.
- Built a robust data pipeline using Spark for processing, Hive for storage, and Kafka for real-time intake.
- Created a Java application listening to Kafka topics, parsing XML files, and storing extracted data in Hive tables in batches of 50,000 records.
- Developed a RESTful API for PayPal Honey Entity using Spring and Java.
Senior Module Lead — Saba
Dec 2017 – Apr 2020 · Pune, India
- Transformed learning module processing with a high-performance Kafka pipeline — converting a serial job system into a parallel pipeline boosted throughput 10x, processing 1 million records in 1 hour.
- Architected a data pipeline integrating archived and current data into a cost-effective Big Data cluster, reducing report generation for 5 million records from 24 hours to 30 minutes.
- Developed new features and REST services for the Certification and Curriculum modules.
Associate (Software Developer) — Deutsche Bank Group, Market Risk IT
May 2016 – Dec 2017 · Pune, India
- Built a robust framework within Matrix that efficiently calculates valuations for a massive portfolio, streamlining the trade valuation process by 50% across 90,000+ listed trades.
- Implemented optimizations using Java, Hazelcast, and multi-threading in a distributed system, increasing trade data processing speed 48x (48 hours to 4 hours).
- Revamped exposure assessment with advanced back-testing methods, resulting in 30% more accurate risk analysis.
- Wrote a Hadoop MapReduce tool processing 1 million trades in HBase in 10 minutes (67% time reduction).
- Increased team productivity by 25%; received Recognition and Spot Excellence awards.
Senior Software Engineer — Zensar
Feb 2015 – May 2016 · Pune, India
- Developed REST APIs for Breeds, Judges, and Events microservices, cutting data access time by 50%.
- Developed secure login services using Spring Security for LDAP, GIGYA, and database systems.
- Automated data synchronization, slashing manual effort by 80% across Oracle and MySQL databases.
Senior Software Engineer — HSBC
May 2011 – Feb 2015 · Pune, India
- Created automated data upload for daily updates of us.hsbc.com customer data, improving customer engagement by 30%; handled customer e-statements and added features to the e-statement system in retail banking.
- Streamlined campaign data retrieval, saving the Content Management team 1 hour per day.
Software Engineer — CDAC R&D
Mar 2010 – Apr 2011 · Pune, India
- Implemented and customized an Account, Budget, Works, and Billing Monitoring System — 20% increase in client satisfaction; custom Billing Module reduced errors by 30%.