Khushi Singh

I’m a data engineering analyst with 3+ years of enterprise experience delivering ETL/ELT pipelines and analytics-ready datasets. At Deloitte US India, I diagnosed and resolved recurring GCP timeout failures by refactoring a BigQuery SQL pipeline, reducing runtime from 6 hours to 25 minutes (93% reduction).

I built and maintained SQL/PLSQL and cloud pipelines across 3+ long-term client engagements, supporting quarterly reporting and downstream business analytics. I also migrated 85+ on-prem dashboards to cloud with zero data loss, retrofitting table schemas per environment requirements and validating end-to-end results.

Beyond engineering, I focus on usable outcomes for teams. I converted legacy outputs into optimized analytical views in the Enterprise Data Hub (EDH), corrected data defects at the view layer, and handed off clean datasets for Tableau-based business metrics consumption—while enforcing data quality with zero critical failures across enterprise reporting datasets.

Currently, as a Research Assistant, I build Python/Pandas ETL pipelines and sensor-vs-baseline comparison workflows, and I deliver multi-tab Streamlit climate analytics dashboards adopted for simulation workflows and EPW file export. Earlier, I also built an Azure-based social ingestion system with Facebook OAuth/Graph API and deployed serverless OpenAI GPT-4 capabilities to power a customer-facing feature with automated PDF export.