Himanshu Sharma

GenAI & Analytics Engineer

Gurugram, Haryana, India 3+ yrs exp 86 · Excellent

About

Analytics and GenAI engineer with 3 years at Volkswagen Global Digital Solutions India (VWITS), building Apache Superset BI tools on Trino/Hive and private, on-premise LLM applications for text-to-SQL. Currently fine-tuning Qwen3-4B locally for Trino SQL generation on sensitive enterprise data. Pursuing an M.Tech in AI & ML at BITS Pilani.

Skills & Expertise (19)

Python Advanced

8.5/10

Years Exp

SQL Advanced

8.0/10

Years Exp

Prompt Engineering Advanced

8.0/10

Years Exp

AI Agents Intermediate

7.5/10

Years Exp

Typescript Intermediate

7.5/10

Years Exp

ReactJS Intermediate

7.5/10

Years Exp

MySql Intermediate

7.0/10

Years Exp

Git Intermediate

7.0/10

Years Exp

Docker Intermediate

7.0/10

Years Exp

JFrog Artifactory Minio Bitbucket ArgoCD Red Hat OpenShift Apache Superset MongoDb Hive Trino JavaScript

Work Experience

Data Visualization & Analytics Developer

Volkswagen Global Digital Solutions India (VWITS)

Aug 2023 - Present

Built a text-to-SQL assistant for the ODA analytics platform (Python, LiteLLM routing to private model endpoints) that lets business users query Trino in plain English; reached ~90% correct SQL on internal test questions. Designed the tool's privacy setup so schema metadata and data are processed only on private enterprise endpoints, with nothing sent to external AI APIs. Built a chart-recommendation module that detects the shape of a query result and suggests a suitable visualization. Built and maintain 15+ production Superset dashboards used by automotive operations teams globally; modular filters cut analysis turnaround time for stakeholders by 40%. Optimized Trino/Hive queries over multi-million-row datasets, improving processing speed by 30% and reducing query timeouts at peak load. Developed custom Apache Superset chart plugins in ReactJS/TypeScript for visualizations native Superset lacks, plus cross-filtering and dashboard state management for self-service reporting. Built REST APIs on MinIO object storage to save, export and share dashboard filter configurations as JSON across user groups. Containerized and deployed services with Docker on Red Hat OpenShift, with continuous delivery through ArgoCD and Bitbucket.