MS Robotics student at Northeastern University, Boston with a focus on 3D vision, deep learning for perception, and autonomous systems. I work at the intersection of classical geometry and modern neural methods β from 3D Gaussian Splatting and Structure-from-Motion to sensor fusion and Vision-Language-Action models.
I believe in building things from scratch first β understanding the math before using the library.
- π MS in Robotics β Northeastern University, Boston (GPA: 3.636)
- π Currently exploring Vision-Language-Action (VLA) models and neural scene representations
- π οΈ Building from scratch: DQN in PyTorch β Gaussian Splatting β fine-tuning VLAs on sim benchmarks
- π Boston, MA
- πΌ Actively seeking full-time roles in Perception & Autonomy
| Project | Description | Stack |
|---|---|---|
| DarkSight | Perception-aware low-light enhancement for autonomous systems β DS3DConv + ConvGRU recurrent pipeline with RAFT temporal stability, TensorRT deployment at 21 FPS, and YOLOv8 integration on Intel RealSense D435 | PyTorch TensorRT Zero-3DCE RAFT YOLOv8 RealSense |
| gaussian-splatting | 3D Gaussian Splatting from scratch β Buddha statue reconstruction with COLMAP, Docker, and CI/CD | Python COLMAP Docker 3DGS |
| 3D-Reconstruction-using-SfM | Incremental SfM pipeline with OpenCV PnP tracking and GTSAM Bundle Adjustment | Python OpenCV GTSAM SfM |
| Semantic-Color-Constancy | Novel color constancy method leveraging object-level semantic priors β CS 7180 Advanced Perception | PyTorch Semantic Segmentation CV |
| Camera-LiDAR-Temporal-Calibration | Temporal calibration between camera and LiDAR for autonomous field robotics | Python ROS Sensor Fusion Calibration |
| 2D-EKF-using-GPS-and-IMU | Extended Kalman Filter fusing high-frequency IMU with async GPS for vehicle trajectory estimation | Python EKF IMU GPS |
| Degree | Institution | GPA | Focus |
|---|---|---|---|
| MS in Robotics (2024 β 2026) | Northeastern University, Boston | 3.636 | 3D Vision, Deep Learning for Robotics, Autonomous Systems |
| BTech in Electronics & Communications Engineering (2020 β 2024) | Vellore Institute of Technology, Chennai, India | 8.09 / 10 (3.48 GPA) | ML, Digital Image Processing, Computer Vision, Deep Learning |
3D Scene Representation β Gaussian Splatting, NeRF, SfM, SLAM
Perception β Color Constancy, Image Enhancement, Sensor Fusion
Embodied AI β Vision-Language-Action Models, Sim-to-Real Transfer
State Estimation β EKF/UKF, Camera-LiDAR Calibration, IMU Fusion
Geometry first. Neural next. Ship when it works.
Ask me about: 3D Gaussian Splatting Β· SfM Β· SLAM Β· Sensor Fusion Β· VLMs Β· Deep RL