Headline

Data Platform Engineer | Kafka • PySpark • Snowflake • AWS

About

Data Platform Engineer with 3+ years of experience building and operating production data systems for connected-vehicle telemetry. I work across the full data lifecycle, from high-frequency Kafka ingestion and Python-based payload decoding to PySpark processing, AWS orchestration, Snowflake modelling, and downstream analytics. At Hero MotoCorp, I reduced AWS Glue infrastructure costs by 30%, designed backward-compatible multi-tenant Snowflake architectures for zero-downtime platform onboarding, and built SLA-driven Step Functions workflows for high-volume telemetry pipelines. I also developed a Python decoding engine that converted raw hexadecimal sensor payloads into analytics-ready data, enabling QA and business teams to investigate vehicle issues without engineering intervention. My current focus is applying this data-platform foundation to production AI systems, including RAG, LLM orchestration, agentic workflows, evaluation, and observability for enterprise data and operational use cases.

Location

Greater Bengaluru Area

AI stack

codex

gemini

github copilot

langchain

langfuse

langgraph

Work experience

Data Engineer at Hero MotoCorp, Apr 2025 - Present · 1 yr 8 mos, Bengaluru, Karnataka, India, - Reduced AWS Glue infrastructure spend by 30% by profiling PySpark execution plans, eliminating redundant shuffles, and right-sizing DPU allocation across production ETL workloads. - Designed backward-compatible, tenant-aware Snowflake stored procedures that enabled zero-downtime onboarding of new ICE and EV vehicle platforms while preserving existing downstream consumers. - Architected multi-stage AWS Step Functions workflows for high-volume IoT telemetry pipelines, enforcing dependency ordering, retry behaviour, failure handling, and SLA-based execution. - Improved Snowflake loading throughput and query performance by migrating high-use transformation paths from external data access patterns to optimised internal staging tables. - Built a modular Python API framework used by five cross-functional teams to automate telemetry extraction, processing more than 10,000 records per execution cycle and eliminating recurring manual analysis. - Implemented tenant-aware Snowflake transformation layers that enforced logical isolation between ICE and EV vehicle data streams. - Partnered with QA, analytics, platform, and vendor teams to investigate telemetry failures, translate operational requirements into data workflows, and resolve production issues across the vehicle-to-cloud pipeline.

Data Engineer | Associate Connected Platform Manager at Hero MotoCorp, Jun 2023 - Mar 2025 · 1 yr 10 mos, Haryana, India, - Built a Python telemetry decoding engine that transformed raw hexadecimal sensor payloads into structured, analytics-ready records, reducing engineering dependency during QA and vehicle issue investigation. - Owned technical integration across mobile applications, cloud infrastructure, and IoT telemetry streams for ICE and EV fleets. - Converted previously opaque binary telemetry into readable diagnostic outputs, enabling QA and business teams to investigate customer-reported vehicle issues more efficiently. - Monitored vehicle-to-cloud telemetry flows, identified stream failures, and coordinated vendor resolution to maintain platform SLA compliance.

Education

Postgraduate Degree, Internet of Things (IOT), Centre for Development of Advanced Computing (C-DAC)

Bachelor of Engineering, Electronics, Government Engineering College Raipur