At Escape™ App AI, I focused on applied NLP and Generative AI in the healthcare sector. My work involved building a FastAPI application for self-care Q&A, integrating advanced LLMs, and deploying services to cloud environments. I also trained MusicGen for domain adaptation in generative audio, enhancing the application's capabilities and user experience.

Worked on applied NLP and LLM research in the healthcare domain, focusing on patient discourse analysis and medical text processing. Built a RAG system using GPT-4o and ChromaDB for contextual Q&A, developed multi-agent pipelines with LangChain to boost medical NER accuracy, and fine-tuned domain-specific models like BioBERT and LLaMA. Leveraged OpenAI embeddings for clustering and deduplication, contributing to high-precision, scalable solutions in medical NLP.

Worked on enterprise ETL and data warehousing across IBM z/OS and Teradata Vantage. Architected COBOL programs orchestrated with JCL, developed DataStage jobs, and wrote Advanced SQL for large-scale data operations and aggregation. Added PySpark backfills on Hadoop with parity checks to Teradata outputs, introduced data quality automation and transaction-level validations for critical pipelines, and supported performance tuning, documentation, and version control within team workflows.

Performed ETL tasks using Teradata SQL and conducted data analysis to create robust data pipelines. Conducted root cause analysis. Worked on Python automation to check the data flow in the HDFS to the Teradata Data warehouses.