Vladislav Bronzov
AI Engineer at Pigment
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolVladislav Bronzov is an AI Engineer based in Paris with six years of experience building back-end systems and machine learning solutions using Python and C#. He has shipped production microservices and desktop apps at Microsoft and Tecan, and now focuses on ML engineering and MLOps practices at Pigment. An active contributor to Hugging Face Transformers, he implemented GGUF support and model loading for major architectures—work that directly improves model interoperability and inference efficiency. Comfortable across cloud environments, container orchestration, PyTorch/ONNX toolchains and distributed training (including FSDP), he bridges research-grade model implementations and robust production pipelines. Vladislav’s background in innovation management and economics underpins a pragmatic approach to turning complex ML research into reliable, maintainable software.
6 years of coding experience
5 years of employment as a software developer
Bachelor's degree, Economics, Bachelor's degree, Economics at Moscow Witte UniversityMaster's degree, Innovation Management, Master's degree, Innovation Management at Finance University under the Government of the Russian Federation
Russian, English
Github Skills (12)
transformers10
quantization10
pytorch10
machine-learning10
gff10
model-conversion10
language-model10
gf10
python10
gfs10
nlp9
deep-learning9
Programming languages (1)
Github contributions (5)
huggingface/transformers
May 2024 - Mar 2025
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
Role in this project:
ML Engineer
ML EngineerContributions:70 reviews, 18 PRs, 175 comments in 10 months
Contributions summary:Vladislav primarily focused on adding and improving GGUF (GGML Unified File Format) support for various language models within the Hugging Face Transformers library. Their contributions included implementing loading support for new architectures like Qwen2Moe, Bloom, StableLM, GPT-2, Starcoder2, and Mamba. They also addressed bugs and refactored code to ensure accurate weight conversion and tensor mapping for these models. Additionally, the user added unit tests to validate GGUF integration.
pythonbertspeech-recognitionstate-of-the-artflax
VladOS95-cyber/transformers
Sep 2024 - Mar 2025
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
Contributions:139 pushes, 14 branches in 6 months
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.