Headline

Machine Learning Engineer

About

I work on applied machine learning and computer vision systems, focusing on turning visual data into usable spatial understanding for real-world environments. At Track3D.ai, I develop and productionize image-to-point-cloud reconstruction and spatial analytics pipelines built on photogrammetry and Structure-from-Motion (SfM). My work involves improving reconstruction robustness, designing automated point cloud segmentation and measurement workflows, and integrating deep learning models into scalable production systems. My experience spans 3D reconstruction, multi-view geometry, feature extraction and matching, depth estimation, and point cloud processing, along with practical deployment of models using PyTorch, Python APIs, and containerized workflows. I have worked extensively with COLMAP-based pipelines, learned feature methods, and computer vision preprocessing systems for large-scale real-world datasets. Alongside computer vision, I explore applied Generative AI and Retrieval-Augmented Generation (RAG) systems, focusing on building practical solutions that bridge machine learning research with production use cases. Always open to conversations around machine learning, computer vision, and building practical AI systems that work beyond the lab.

Location

Hyderabad

AI stack

claude code

codex

cursor

hugging face

openai

huggingface

Work experience

Machine Learning Engineer at Nasuni, Apr 2026 - Present · 7 mos, Hyderabad

Machine Learning Engineer at Track3D, Jun 2023 - Apr 2026 · 2 yrs 11 mos, On-site

Machine Learning Intern at Track3D, Dec 2022 - Jun 2023 · 7 mos, Hyderabad, Telangana, India

Education

Master of Computer Applications - MCA, PSG College of Technology

Bachelor's degree, Computer Science, Grd College of Science

Public links

github: https://github.com/kishoreprasad