Georgi Tinchev
Applied Scientist at Amazon
London, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolGeorgi Tinchev is an applied scientist in London with 11 years of experience building perception and ML systems, currently advancing 3D perception and computer vision at Amazon. He holds a DPhil from Oxford and has a track record of translating research into production—ranging from neural radiance field prototypes and correspondence networks to large-scale text-to-speech models. His background spans robotics, LiDAR and point-cloud processing, and scalable deep learning implementations in C++, TensorFlow and PyTorch, with practical deployment on HPC, Hadoop and AWS clusters. Notably, he has led end-to-end localization and multirobot perception work on platforms like NASA Valkyrie and ANYmal, blending statistical methods with modern feature-learning approaches. Colleagues value his ability to compress complex experimental pipelines into efficient, production-ready solutions.
11 years of coding experience
8 years of employment as a software developer
Bachelor of Science (BSc), Computer Science, 1st Class Honours (4.0 GPA), Bachelor of Science (BSc), Computer Science, 1st Class Honours (4.0 GPA) at University of AberdeenMaster of Science (MSc), Artificial Intelligence, 3.71 GPA, Master of Science (MSc), Artificial Intelligence, 3.71 GPA at The University of EdinburghDoctor of Philosophy (DPhil), Engineering Science, Doctor of Philosophy (DPhil), Engineering Science at University of Oxford"Baba Tonka" High School of Mathematics
English, Russian, Bulgarian, German
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (64)
generative-model10
correspondence9
pcl9
python9
machine-learning9
deep-learning9
slam-algorithms9
orb9
cpp8
transfer-learning8
computer-vision8
point-cloud8
robot8
tensorflow8
diffusion-models8
Programming languages (6)
Github contributions (5)
gtinchev/3DFeatNet
Mar 2019 - Dec 2023
3DFeat-Net: Weakly Supervised Local 3D Features for Point Cloud Registration
Contributions:2 PRs, 1 push, 2 branches in 4 years 10 months
3d-modelssupervisedpoint-cloud-registrationregistrationvolume-rendering
XYZ-R-D/xrcnet
Dec 2020 - Oct 2021
A novel dense correspondence network
Contributions:21 commits, 2 PRs, 21 pushes in 10 months
correspondencedense
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.