Rahat Khan

I’m a Machine Learning Engineer and Data Scientist who enjoys turning unstructured data into useful, retrieval-driven experiences. I’ve focused on NLP workflows, from preparing knowledge for LLM retrieval to validating models with practical cross-validation.

In my ML Engineer Intern—OJT experience, I contributed to a Retrieval Augmented Generation (RAG) development pipeline. I processed and structured ML training content by creating transcripts and organising knowledge chunks in Excel for RAG system ingestion, gaining hands-on exposure to RAG architecture and how unstructured content is prepared for LLM-based retrieval systems.

My projects reflect how I build end-to-end solutions: ForensiQ used an OCR pipeline, spaCy NER, and TF-IDF similarity with a secured Flask REST API (JWT + role-based access) and PDF exports; I also deployed models as interactive Gradio/Streamlit apps (including an IPL match winner predictor and crime type prediction). I’m looking for an environment where I can keep shipping applied ML—especially NLP, RAG, and clean deployment workflows.