AI/ML Developer Intern at Capital One

Improved RAG retrieval accuracy by 32% and cut search latency by 45%, enabling faster and more reliable insights. Built large-scale pipelines processing 2.5B+ records, improving data quality and model performance end-to-end. Automated data labeling workflows, decreasing manual effort by 60% while maintaining high precision. Achieved 99.8% pipeline uptime and saved 8+ TB/month in storage through optimized curation and indexing. Accelerated experiment cycles 3× and supported 12+ ML teams with clean, production-ready datasets.

Data Engineer at Infosys (Client: Wells Fargo)

Designed scalable data pipelines on AWS using Glue, Lambda, Step Functions, and S3 to support high volume ingestion and transformation across multiple business units. Developed optimized ELT workflows with Spark and Snowflake, improving query performance and reducing processing costs through clustering, materialized views, and task based scheduling. Built reusable data quality frameworks with Python, SQL, and PySpark to validate schema, lineage, and SLAs, increasing pipeline reliability and reducing production defects. Created near real time streaming pipelines with Kafka and Kinesis to support event driven applications and operational dashboards. Implemented CI and CD automation with GitHub Actions and Terraform to standardize infrastructure deployments and ensure consistent releases across staging and production. Collaborated with cross functional teams to translate data requirements into scalable models and governed datasets, enabling faster reporting and advanced analytics use case