I'm a third-year AI/ML student at GITAM (Deemed to be University), curious about what happens beyond the abstractions we use every day. I volunteer as a tester for DeepLearning.AI courses, helping identify issues, inconsistencies, and areas for improvement before courses are released. I'm passionate about understanding how things actually work under the hood, not just using libraries and frameworks. I enjoy exploring the underlying concepts, experimenting with implementations, and learning by building. I'm looking to collaborate with people who enjoy reverse-engineering how applications work, building interesting products, or exploring ideas beyond the surface level. If you're building an application and need someone to work on the ML, AI, or agentic systems side of things, I'd love to connect and build something together.
Volunteer Tester
Deeplearning.ai
- I volunteer in testing upcoming AI courses
A mini assignment project that was written as a curiosity to understand how computers communicate with each other with a client-server architecture
Built an end-to-end ML pipeline on the Ames Housing dataset to predict home sale prices. It covers EDA, custom feature engineering and preprocessing, model comparison with 5-fold CV, and hyperparameter tuning. The final tuned Extra Trees model reached an R² of 0.927 and a MAE of about $14.8k on a held-out test set.
Built a passenger satisfaction classifier for Kaggle Playground Series S6E10. I did leakage-aware feature engineering, then trained and tuned an MLP, LightGBM and CatBoost, and combined them with a rank-averaged blend for the final submission. Earlier experiments reached a holdout ROC-AUC of about 0.956.
Protégé is an “AI student” learning app where you learn by teaching, Feynman-style. It acts as a curious Gemini-powered student that asks follow-up questions on text, voice, or diagram input. A multi-agent pipeline built on Google ADK then scores your explanation and produces a mastery report.
B.Tech
GITAM deemed to be University, Hyderabad · 8.6
Deeplearning.ai
Jun 2026Deeplearning.ai
Oct 2026
Attendee · Oct 24, 2026