I'm a Robotics Engineer specializing in computer vision, deep learning, and perception for embodied AI. Currently pursuing my Robotics MSE at Johns Hopkins University, I focus on building intelligent systems that bridge vision, language, and action for autonomous robots.

​

My work spans visual perception, multimodal AI, and learning-based autonomy. Recent projects include developing neural image enhancement pipelines for robust feature matching (34% improvement over baseline), implementing visual SLAM for localization, and training CNN-based detection models for obstacle avoidance. I'm currently exploring vision-language-action models and reasoning architectures for embodied systems, working toward thesis research on grounding language models in physical robot interactions.

​

Technical Focus Areas:
- Vision-Language-Action Models: Multimodal architectures, reasoning for embodied AI, vision-language grounding.
- Computer Vision: Visual SLAM (ORB-SLAM), feature matching (SIFT, SuperPoint), image denoising, camera geometry.
- Deep Learning: PyTorch, transformer architectures (Restormer, VLMs), CNNs (YOLO, Faster R-CNN), model fine-tuning and post-training (SFT, RLHF).
- Perception for Embodied AI: Multimodal perception, sensor fusion, visual odometry, 3D understanding
- Foundation Models: Diffusion models, generative approaches for robotics, scene understanding and manipulation.
- Robotics Systems: ROS2, path planning (RL-based, classical), autonomous navigation, sim-to-real transfer.

​

I enjoy tackling problems at the intersection of multimodal AI and physical robotics—whether that's training vision-language models for robot instruction following, fine-tuning transformers for visual reasoning, or designing perception pipelines that enable real-world autonomy. My experience spans both algorithm development (VLA architectures, generative models) and deployment (embedded vision systems on UAV hardware), giving me perspective on moving from research ideas to production systems.

​

Published Research: VQ-VAE-based trajectory generation for multi-drone coordination (ICCA 2025).

​

Currently seeking 2026 summer research internships in multimodal AI, computer vision, and embodied intelligence where I can contribute to vision-language-action models, foundation models for robotics, or learning-based perception systems.

​

Open to opportunities in: Vision-language models, multimodal reasoning, VLAs for agentic systems, generative models for robotics, perception for autonomous systems, and applied ML for embodied AI.