Georgiย Gerganov
Software Engineer at Hugging Face
Sofia, Sofia-City, Bulgaria
Summary
๐คฉ
Rockstar๐
Top SchoolGeorgi Gerganov is a software engineer with nine years of experience building high-performance C/C++ systems for ML inference, audio networking, and real-time UIs, now working at Hugging Face. He founded and led ggml to productionize edge AI and is the primary author of notable open-source projects like whisper.cpp and wave-share, demonstrating deep expertise in model conversion, quantization, and audio-based data transfer. His background as a Principal Scientist in medical physics informs a rigorous, performance-oriented approach to system design and numerical correctness. Comfortable across the stack, he contributes to both low-level optimization (BLAS/threading, kernel fixes) and front-end UX refinements, and has a playful, productivity-focused developer ethos (big .vimrc).
9 years of coding experience
9 years of employment as a software developer
Master of Science (MSc) Medical Physics, Master of Science (MSc) Medical Physics at Sofia University St. Kliment Ohridski
Stackoverflow
1,036reputation
46kreached
21answers
2questions
Github Skills (52)
algorithm10
fftw10
algorithms10
compatible10
javascript10
c-language10
python10
machine-learning-models10
algebra10
sdl10
css10
machine-learning10
user-interface10
c1110
blas10
Programming languages (15)
Github contributions (5)
ggml-org/ggml
Sep 2022 - Jan 2023
Tensor library for machine learning
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:349 reviews, 44 commits, 481 PRs in 4 months
Contributions summary:Georgi contributed to improving the performance and functionality of the machine learning library by adding support for runtime performance timings and optimizing the code. The commits also included modifying thread counts and example files and adding new components to expand the functionality of the library. The main contributions focus on improving the overall performance of the library.
deep-learningllmmachine-learningtensor-libraryneural-network
ggerganov/kbd-audio
May 2018 - Jan 2023
๐คโจ๏ธ Acoustic keyboard eavesdropping
Role in this project:Back-end Developer & System Architect
Back-end Developer & System ArchitectContributions:2 releases, 213 commits, 22 PRs in 4 years 8 months
Contributions summary:Georgi was instrumental in developing the core functionalities of this acoustic keyboard eavesdropping project. They refactored key logging into a class, implemented an audio logger to capture data, and added functionality to play back recorded audio. Their work included the creation of essential components and the integration of diverse technologies for audio capture, indicating their involvement in both back-end development and architecting the system's core functionality. The user also created tools for recording and playback, and improved the overall process.
acousticmidimicrophone-audio-capturekeyboardeavesdrop
Find and Hire Top DevelopersWeโve 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.