Anatoly Belikov
Research Scientist at SingularityNET
Dubai, Dubai, United Arab Emirates
Summary
👤
Senior🎓
Top SchoolAnatoly Belikov is a research scientist and seasoned software engineer with 12 years of experience building cloud platforms, distributed systems, and ML tooling from Dubai. He currently contributes to SingularityNET and is an active open-source contributor to high-profile projects such as Hugging Face Diffusers—where he implemented masked image-to-image and inpainting pipelines for Stable Diffusion—and to the OpenCog Atomspace graph system, optimizing core hashing and graph handling. His background spans performance-focused backend engineering, benchmarking distributed filesystems, and test automation, reflecting a rare blend of research rigor and production-grade engineering. Holding advanced degrees in computer science and computer engineering from top Russian institutions, he pairs academic publication activity with hands-on system design. Colleagues describe him as the kind of engineer who moves fluid research ideas into robust, reproducible code and training pipelines.
12 years of coding experience
6 years of employment as a software developer
Engineer's Degree, Computer Engineering, Engineer's Degree, Computer Engineering at Moscow State University of Instrument Engineering and Computer Sciences (MGUPI)Master's Degree, Computer Science, Master's Degree, Computer Science at Bauman Moscow State Technical University
Github Skills (29)
pytorch10
c-language10
python10
atomics10
diffusion-models10
machine-learning10
stable-diffusion10
graph-database10
diffusers10
diffusion-probabilistic-models10
text2image10
knowledge-representation10
huggingface-transformers10
diffusion-probabilistic10
deep-learning10
Programming languages (15)
Github contributions (5)
opencog/atomspace
Apr 2018 - Apr 2021
The OpenCog (hyper-)graph database and graph rewriting system
Role in this project:
Back-end Developer
Back-end DeveloperContributions:146 commits, 35 PRs, 239 comments in 3 years
Contributions summary:Anatoly primarily focused on improving the Atomspace database and graph rewriting system. Their contributions included implementing a new hash function (FNV1a) for the `ScopeLink` class to optimize performance and correct handling of unordered links. The user also refactored code, moved hashing functions to a separate header file and improved the hashing algorithms by taking bigger chunks of data. They introduced improvements to the hash computation, by incorporating arity and interval maps for optimization.
logic-programmingopencogknowledge-baseedgedbknowledge-graph
huggingface/diffusers
Jul 2023 - Aug 2024
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.
Role in this project:ML Engineer
ML EngineerContributions:5 reviews, 12 PRs, 59 comments in 1 year 1 month
Contributions summary:Anatoly primarily contributed to the `diffusers` repository, focusing on implementing and adapting image inpainting and training functionalities within the context of Stable Diffusion models. Their work included creating a masked image-to-image pipeline for non-inpainting models, adapting the masked inpainting pipeline for SDXL, and incorporating loss logging features for training scripts. The user also addressed model training scripts, including handling LoRA scale and CLIP skipping in long prompt weighting pipelines, and modifying image processing behaviors.
pytorchartdeep-learningimage2imagestate-of-the-art
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.