Headline

Software Engineer 3 @ Wayfair | Java, Kafka, Event-Driven

About

I design and own production systems that must stay correct, low-latency, and evolvable under real-world scale. My work focuses on system architecture and design — taking ambiguous problem statements and turning them into clear technical decisions, resilient architectures, and production-ready platforms. At Wayfair, I’ve led the design of event-driven systems processing 300M+ events/day, enabling real-time synchronization across services and reducing data staleness from hours to seconds, while maintaining sub-100ms processing latency. Across e-commerce platforms, master/reference data systems, and data protection infrastructure, I consistently work on problems where correctness, failure handling, observability, and long-term maintainability matter more than quick fixes. I’m particularly drawn to systems that: - operate at scale under tight latency constraints - evolve over time without breaking downstream consumers - require careful tradeoffs across consistency, performance, and cost If you’re building a system where design decisions matter as much as code, I’d be happy to connect.

Location

Bangalore Urban

AI stack

claude code

cursor

Devin

Work experience

Software engineer 3 at Wayfair, Aug 2025 - Present · 1 yr 2 mos, Bengaluru, Karnataka, India, - Led the end-to-end design and delivery of the Supplier Data Cache, a low-latency read platform powering Unified Product Listing experiences across catalog health, findability, recommendations, revenue insights and exports for Wayfair’s 130M+ product catalog. - Re-architected supplier catalog access from high-latency, multi-domain synchronous retrieval to asynchronous Kafka-driven projections, scaling the platform to process 300M+ catalog events per day and serve 10M+ daily reads. - Built a production-grade Apache Flink catalog aggregation pipeline that consumes Bigtable CDC and refresh events from Google Cloud Pub/Sub, hydrates current catalog state from Bigtable, expands market context and publishes consolidated Avro events to Kafka. - Designed an asynchronous bulk export/import platform supporting 100K–250K-row, multi-sheet Excel workflows across multiple catalog domains. Decoupled request handling from workbook generation to meet a 1-second request-acceptance SLA. - Developed a bounded-concurrency orchestration layer that correlates data from independent domain services using parallel execution, rate limiting, timeouts and failure isolation while protecting upstream systems. - Designed a catalog-health intelligence initiative that identifies high-impact products, prioritises supplier issues using catalog and business signals, and uses RAG-based retrieval to generate grounded explanations and remediation guidance. - Built production reliability guardrails including idempotent event processing, replay and backfill workflows, Avro compatibility gates, DLQ/DLT handling, content-hash deduplication, consumer-lag SLOs, drift detection, Flink checkpoints and Datadog alerting. Technologies: Java | Spring Boot | Apache Flink | Kafka | Kafka Streams | Google Cloud Pub/Sub | Google Cloud Bigtable | Avro | Protobuf | Schema Registry | Event-Driven Architecture | Microservices | Aerospike | Elasticsearch | PostgreSQL | GCP | GKE | Datadog | GraphQL

Senior Software Engineer at Informatica, Apr 2022 - Aug 2025 · 3 yrs 5 mos, Bengaluru, Karnataka, India, - Senior Software Engineer on Informatica’s Reference 360 (cloud-native Master Data / Reference Data Management), building scalable microservices and data processing workflows for enterprise-grade governance and reference data operations. - Designed and evolved cloud-native microservices and data workflows for enterprise-scale Master & Reference Data Management (Reference 360), with a focus on data correctness, governance, and scalability. - Removed a major tenant provisioning bottleneck by redesigning MongoDB write paths with a reactive, parallelized pipeline, reducing CodeValues creation time by ~50% and materially improving org onboarding SLAs. - Built a reusable Spark-based data correction framework to safely traverse and repair master and cross-reference data at scale, enabling automated remediation across multiple codelists and large datasets.

Software Engineer at Dell EMC, Feb 2020 - Apr 2022 · 2 yrs 3 mos, Bengaluru, Karnataka, India, - Software Engineer 1 on a NAS Data Protection platform, building backend services and frameworks for discovery, backup, and restore workflows across enterprise storage environments. - Developed a NAS discovery feature to detect and catalog NFS/SMB shares and index discovered assets into Elasticsearch, enabling faster search, tracking, and operational visibility. - Built a Java-based backup/restore framework for filesystem protection, including orchestration, validations, and error handling for reliable operations at scale. - Improved backup throughput by ~30% by segmenting large shares into ~200GB chunks and implementing multi-threaded backup execution, reducing wall-clock backup time and increasing parallelism. - Implemented restore capabilities including full restore and file-level restore (FLR), and enabled cross-storage array restores (e.g., Unity SMB → PowerStore SMB) while enforcing compatibility constraints (e.g., blocking unsupported SMB↔NFS transfers based on platform rules).

Software Engineer at Epsilon India, Jun 2019 - Feb 2020 · 9 mos, Bengaluru South, Karnataka, India

Software Engineer Intern at WorkOnGrid, Feb 2019 - Jul 2019 · 6 mos, Bengaluru, Karnataka, India

Education

Bachelor of Technology - BTech, Computer Science, Visvesvaraya Technological University