Resume

• Software internship on an AI and Python stack, working up from the language itself to retrieval-augmented generation.
• Began in core Python, then moved to FastAPI to build services.
• Backed those services with PostgreSQL and designed the REST APIs on top of them.
• Now working on RAG: retrieval pipelines that answer from a chosen set of documents rather than from the model alone.
• Built docs-rag, a strictly grounded RAG pipeline that answers only from the documents it is given and cites the passages it used.

Built bidirectional cSpell dictionary synchronization extension using TypeScript and VS Code API with 2,100+ installs.

Recognized for developing an innovative web application based on Facial Emotion Detection using MediaPipe and machine learning, showcasing technical expertise and problem-solving skills in a competitive environment

Outstanding performance in delivering high-quality software solutions for real-world scenarios, and utilizing advanced technologies like blockchain and AI to create impactful applications

1st Runner-up at Code Kurukshetra with MomCare — an end-to-end pregnancy companion app covering the first trimester through birth and the newborn's early weeks.

Achieved the top score in a hackathon focused on Pathway.com's stream-processing framework, demonstrating rapid learning and mastery of cutting-edge real-time data engineering technology.

Won the campus round of Microsoft's globally recognized Imagine Cup by building MomCare — a complete end-to-end pregnancy companion app covering everything from the first trimester through birth and the newborn's early weeks.

Due to its high morbidity and medical expenses, DR mellitus (T2DM) is a serious global health concern. Its burden can be lessened with effective early prognosis, but the available treatments are not enough. Predictive modelling of Retina Images (fundus) can enhance early diagnosis and care quality. Advanced convolutional neural networks (Inception-V4, ResNet V1, and ResNet V2) are used in this study to assess the severity of Diabetic Retinopathy caused by T2DM. Using de-identified fundus data from 9,948 individuals (1,904 with a T2DM diagnosed), we used transfer learning to train these models. We have performed simple data augmentation to rectify the class imbalance. The accuracy and AUC of the ensemble model were 94.21% and 98.53%, respectively. The findings imply that the combination of ResNet and Inception-V4 enhances classification power, providing a viable approach for early T2DM identification from fundus data that may find use in clinical settings. Future research still needs to focus on improving sensitivity and integrating real-time.

The EquiLens Corpus Generator is an open-source tool for creating balanced, reproducible datasets for bias auditing in Small Language Models (SLMs).

• Serve as the elected class advisor for the B.Tech CS batch, acting as the primary liaison between students and faculty.
• Coordinate academic schedules and extracurricular activities, address student concerns, and help maintain a healthy, collaborative class environment.

• Participate in robotics challenges and workshops, contributing to hardware-software integration projects.
• Collaborate with team members to design and prototype autonomous systems combining sensors, actuators, and custom control logic.

• Engage in ACM-organized technical talks, coding events, and competitive programming sessions.
• Collaborate on inter-chapter hackathons and stay current with emerging CS research through the ACM Digital Library.

• Maintain 25+ open-source repositories under the Life-Experimentalist GitHub organization, including tools with active community usage.
• Review pull requests, triage issues, write documentation, and mentor contributors on projects ranging from VS Code extensions to AI research tooling.