Abhishek Yadav

Developed internal APIs and ETL services using Python, SQLAlchemy, and PostgreSQL, improving data ingestion efficiency by 30% and supporting ML model testing workflows. Collaborated with engineers and project leads to optimize RESTful endpoints, improving latency by 22% and ensuring consistent performance under concurrent request loads.

Built Power Automate data pipelines and Python scripts to unify outreach datasets across 5+ programs, reducing manual consolidation efforts by 32% and improving data freshness. Prototyped a generative AI membership assistant using Azure OpenAI and custom embeddings, improving response turnaround by 35% while decreasing manual workloads for staff.

Engineered ETL workflows in Python, AWS S3, and PySpark while collaborating with data scientists to process regulatory records, improving downstream AI model readiness by 33%. Built NLP pipelines using TensorFlow and scikit-learn to extract control gaps from supervisory documents, enhancing compliance classification accuracy by 29% for enterprise clients. Optimized semantic similarity models with vector embeddings, reducing inference latency by 24% and improving risk scoring precision for audit teams.