I've built end-to-end RAG applications using LLMs, LangChain, ChromaDB, FastAPI, and Python to support semantic retrieval and context-aware question answering. My work streamlines document ingestion, chunking, and embedding workflows to improve response precision and reduce hallucinations.
During my Machine Learning internship at Cognifyz, I developed regression, classification, and content-based recommendation systems using Python, Pandas, NumPy, and scikit-learn. I worked across preprocessing, exploratory data analysis, feature engineering, model evaluation, and personalized recommendations based on user preferences, cuisine, and price range.
I've also designed a serverless AWS data pipeline with S3, AWS Glue, PySpark, Redshift, IAM, and CloudWatch, delivering clean structured datasets for reporting. In e-commerce customer segmentation, I used K-Means clustering, feature scaling, the Elbow Method, and Silhouette Score to uncover purchasing patterns for targeted marketing.
I enjoy sharing practical machine learning knowledge with the developer community, including delivering a hands-on GDG session on machine learning fundamentals, preprocessing, workflows, and evaluation techniques.
