Chemical Engineering undergraduate at IIT Roorkee.
Core Member β Data Science Group (DSG), exploring ideas where mathematics meets intelligent systems.
Exploring Artificial Intelligence, Deep Learning, and how intelligence can be designed, not just trained.
I care more about understanding concepts deeply than rushing through trends.
Building quietly with Python, PyTorch, and experimental ideas that evolve over time.
Still early in the journey β focused on growth, clarity, and long-term mastery.
| Artificial Intelligence | Languages | Dev & Deployment |
|---|---|---|
    |
    |
    |
Featured Project
Pixel-Play-26-VAD
Memory-Augmented Video Anomaly Detection built on VideoPatchCore.
A research-driven computer vision pipeline developed for the VLG Recruitment Challenge, focused on detecting abnormal events in surveillance footage under noisy and imperfect data conditions.
Instead of relying on heavy generative models, the system uses a memory-based VideoPatchCore architecture with a frozen CLIP encoder to learn representations of normality and localize deviations efficiently.
- Building and experimenting with video understanding & anomaly detection systems.
- Diving deeper into deep learning architectures and representation learning.
- Refining how I design, test, and iterate on AI systems β not just training models.
- Experimenting with web and app development.
- Learning Linux and Bash.
- Balancing engineering fundamentals with long-term AI research interests.
