Profile
Computer vision and multimodal AI, with an end-to-end engineering range.
I work across multimodal generative models, physical AI, and
geometric computer vision. My role typically spans research design,
data and training systems, evaluation, optimization, and product
integration. I am most interested in spatial systems that must
remain reliable under real data, hardware, and infrastructure
constraints.
Current focus Multimodal post-training, real-to-sim, digital twins, and robot calibration.
Research foundation Robust and efficient 3D reconstruction, neural rendering, and multiview geometry.
Technical scope
Methods and systems
- Multimodal ML
- Post-training data, VLM curation, diffusion and flow matching, image and video generation, editing, spatial control, evaluation, and ablation design.
- Geometry
- Robot and camera calibration, real-to-sim, SfM, bundle adjustment, MVS, stereo, RGB-D, SLAM, NeRF, Gaussian Splatting, and 3D/4D reconstruction.
- ML systems
- PyTorch, Ray, vLLM, DDP, FSDP, tensor and context parallelism, quantization, runtime profiling, and distributed model serving.
- Engineering
- Python, C/C++, CUDA, OpenGL/WebGL, AWS, GCP, Docker, backend systems, databases, and interactive 3D applications.
Research
Selected publications
Representative first-authored and collaborative work.
Complete list on Google Scholar ↗
Browse the publication list on
Google Scholar.