Built a mobile app from scratch, streamlining workflows for 1,000+ employees and saving the company $1M annually. Achieved a 95% crash-free rate and boosted performance by 30%.
A mood-based playlist generator powered by LangChain, LlamaIndex, and GPT-3.5, it semantically searches a dataset of 10K+ songs to deliver personalized recommendations that feel like chatting with a DJ.
Face-recognition-based attendance system with 98% accuracy. Integrated with Firebase for secure storage of data and logs, real-time automation. Marks attendance in less than 2 seconds.
Full-stack bookstore platform managing 1,000+ books and 200 daily users. Optimized APIs for 20% faster responses and cut load times by 40%.
Machine learning model with 98% accuracy to detect phishing websites. Published research in IJFMR, validating real-world effectiveness.
Sports platform with CRUD operations and real-time score integration. Increased user engagement by 30% and cut backend response time by 15%.
Predictive model with 95% accuracy to estimate graduate admission chances. Packaged with a simple GUI for easy student use.