About

I am a third-year Computer Science student at PES University.

My work sits at the intersection of systems programming, distributed infrastructure, and machine learning. I am drawn to problems where correctness and performance are both required. I have a particular interest in multi-modal machine learning, and its applications to video understanding and anomaly detection.

Experience

Research Intern — C-ISFCR, PES University

Jun 2025 – Jul 2025, under Prof. Preet Kanwal

Designed and evaluated four multi-modal architectures (YOLOv11 + Optical Flow + LLaVA-Next + LLaMA-3) for unsupervised video anomaly detection, achieving 75.48% classification accuracy on the NWPU Campus Dataset across 16.5 hours of footage.

Built a novel keyword-free composite anomaly scoring framework combining semantic (SBERT), visual (ResNet-50), object distribution (Jensen-Shannon divergence), and temporal dynamics signals. F1 scores of 0.845–0.895 across all test categories. Benchmarked against Gemini-2.5-Flash — comparable anomaly scores at significantly lower serving cost.

Teaching Assistant — Dept. of Computer Science and Engineering, PES University

Jan 2026 – Present

TA for Microprocessors and Computer Architecture (UE24CS351B) under Prof. Chitra GM. Prepared course material and the evaluation schema for the course project, serving 700+ students.

Skills

Distributed Systems Engineering

Rust, C/C++, Go, Python, Kafka, Redis, Linux, Bash

ML Systems and Applied Research

PyTorch, CUDA, OpenCV, Sentence-BERT, YOLOv11, LLaVA, LLaMA, Flower

Infrastructure and Delivery

Docker, Kubernetes, GitHub Actions (CI/CD), Spark, Git