Curriculum Vitae

Software Engineer - Perception

Seneca

Building the perception stack for autonomous drones — automated data annotation and training pipelines, custom depth models, and LiDAR-based detection deployed on the edge.

• Built an automated data annotation and training pipeline by adapting SAM3 and FoundationStereo, used to train YOLO and custom depth models for drone-specific scenarios
• Designed and deployed an end-to-end fine-tuning pipeline for LiDAR-based human detection on AGX-Orin

Machine Learning Engineer, Computer Vision

Meta - Reality Labs

Computer vision for AR/VR perception — segmentation, stereo depth, synthetic data generation, and 3D scene understanding.

• Optimized a hand-object segmentation pipeline, improving GPU utilization by 75%, and integrated SAM3 with box + text prompting into an auto-labeling and QA workflow
• Built end-to-end stereo depth estimation pipelines using a customized NVIDIA Foundation Stereo model for real-world AR/VR perception workloads
• Developed a synthetic data generation pipeline from low-quality multi-view AR/VR captures using Gaussian Splatting (2D/3DGS), evaluating 4DGS variants for fidelity and temporal consistency on a CVPR-scale dataset
• Fine-tuned a language-grounded 3D-VLM for scene understanding and geometry-aware AR/VR panel placement, delivering results on 360K samples

Geometric Computer Vision

Vimaan Robotics

Detection, segmentation, and camera calibration for warehouse inventory automation.

• Deployed a transformer-based (DETR) detection and segmentation system on cloud for pallet/ground recognition, improving mAP50-95 by 5% by customizing decoder outputs
• Developed a camera calibration module with noise modeling, achieving pose estimation accuracy within 0.25 degrees and 2 cm

Engineer

Larsen & Toubro Infotech

Master of Science, Electrical and Computer Engineering

University of California, San Diego

Robotics and AI

• Statistical Learning
• Computer Vision
• Visual Learning
• Motion Planning
• Robot Manipulation

Bachelor of Technology

Indian Institute of Technology, Tirupati

Electrical Engineering

Ego-1K: A Large-Scale Multiview Video Dataset for Egocentric Vision

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

3D Scene Rendering with Multimodal Gaussian Splatting

Asilomar Conference on Signals, Systems, and Computers (ACSSC)

ActiveInitSplat: How Active Image Selection Helps Gaussian Splatting

arXiv preprint

Fog-based Distributed Camera Network system for Surveillance Applications

IEEE International Conference on Robotics and Biomimetics (ROBIO)