Pushpneet Singh

Boston, US (Remote)

• Optimized compute costs by refactoring ETL pipelines supporting AI feature-engineering workflows.
• Migrated legacy bid-strategy optimization jobs for Ads using industry best practices, improving overall efficiency by ~40%.
• Built a standardized data-consumption layer within a medallion architecture to unify access for AI and BI teams.
• Leveraged Databricks Unity Catalog for centralized data governance and eliminated reliance on DBFS-backed tables.

Gurugram

• Designed and developed an in-house no-code middleware tool to configure and automate ETL from Emails, APIs, SAP, PubSub, and Databases.
• Implemented SAP S/4HANA accounting integration covering 97% of accounting volume for real-time financial reporting.
• Integrated third-party APIs to automatically generate and trigger Airflow DAGs.
• Utilized Airflow, Spanner, GCS, GKS, FastAPI, Temporal, Redis, and Snowflake.
• Automated tasks with finance team saving 500 hrs/month.

Contract (6 months)

• Wrote Python scripts from scratch to ingest data from different wearable sources and devices to Data Warehouse which resulted in overall time reduction from 24hrs to 2hrs.
• Utilized Apache Kafka, Apache Druid & Apache Superset to get customized health reports in real time.
• Built trigger based data ingestion pipeline using Azure Function to ingest data from health devices.

Delhi

• Migrated complex on-premise contact center solution (100+ tables) to Cloud Environment for Commonwealth Bank of Australia.
• Developed robust ETL pipelines processing data from 5000+ agents, 6M+ inbound calls, and 1.5M+ digital chats monthly.
• Automated stream job monitoring reducing manual effort by 80%.
• Optimized deployed pipelines to improve daily processing time.