Hi! I'm Talha. I'm a PhD student in Computer Science and Fulbright Scholar at Purdue. I currently research Diffusion Models and Implicit Neural Representations for Grasp and Motion Optimization.

Previously, I was a ML Software Engineer at Pointivo where I developed and deployed ML pipelines to derive insights for 3D asset inspection. Before that, I studied Mechatronics & Control Engineering at UET Lahore, and Signal Processing & Machine Learning at LUMS.

Research Interests

Machine Learning ∪ Motion Planning ∪ Geometry Processing

Selected Publications

2025-vsigd
Variational Shape Inference for Grasp Diffusion on SE(3)
ST Bukhari, Kaivalya Agrawal, Zachary Kingston, and Aniket Bera
(under review); 2025
arXiv
2025-neural-sdf
Differentiable Composite Neural Signed Distance Fields for Robot Navitation in Dynamic Indoor Environments
ST Bukhari, Daniel Lawson, and Ahmed H. Qureshi
IEEE International Conference on Robotics and Automation (ICRA); 2025
arXiv GitHub Webpage
2021-brats-e1d3
E₁D₃ U-Net for Brain Tumor Segmentation
ST Bukhari and Hassan Mohy-ud-Din
International MICCAI Brainlesion Workshop (BrainLes); 2021
arXiv GitHub
2021-pmb-lr
A Systematic Evaluation of Learning Rate Policies in Training CNNs for Brain Tumor Segmentation
ST Bukhari and Hassan Mohy-ud-Din
Physics in Medicine & Biology (IF:3.61); 2021
Paper