MUHAMMED SAFWAN K S

At TELUS International AI, I evaluate search, geographic, audio, image, and labeled data against detailed quality guidelines. I assess AI-related content for accuracy, relevance, consistency, completeness, and sensitive-content criteria.

Previously at Luminar Technolab, I cleaned, validated, transformed, and analyzed structured and semi-structured data using Python, SQL, ETL workflows, Pandas, and NumPy. I also built machine learning models, created Power BI dashboards, and documented experiments and technical workflows.

I’ve built an LLM evaluation and benchmarking platform that analyzes factual accuracy, relevance, instruction following, logical consistency, hallucinations, and safety. I also developed a RAG-based research assistant using document processing, embeddings, vector search, and context-based response generation.

My computer vision projects include CNN-LSTM lip-reading and CNN-based sign-language recognition systems. I bring hands-on experience in data quality, model evaluation, annotation, research, and analytical problem-solving.