AI Intern

What I Did

• ▸Built end-to-end production ML pipelines for disease classification and real estate valuation, achieving 85% model accuracy using supervised learning techniques and automated workflows from data ingestion through deployment
• ▸Implemented full pipeline using TensorFlow, Scikit-learn, and FastAPI — from raw data to deployed model
• ▸Contributed to the Color Analysis project using OpenCV implementation techniques under direct mentorship

Reflection

This is where AI clicked for me beyond theory. Seeing models work on real data, fail in unexpected ways, and improving them systematically, basically I learnt the applications in more depth. Hitting 85% accuracy on a disease classification task as an intern felt like proof that I could build things that actually matter.