Matvey Ezhov
Co-founder & CTO at Diagnocat
Dubai, Dubai, United Arab Emirates
Summary
👤
SeniorMatvey Ezhov is a technology leader and co-founder/CTO with 13 years of experience building production-grade AI systems and R&D teams, currently leading Diagnocat’s next‑gen dental diagnostic assistant from Dubai. He combines hands-on data science, software architecture and backend engineering—comfortable moving between prototype research and scalable streaming systems that handle terabytes per day. At Ostrovok.ru he built an R&D division and shipped high-impact ML products (pricing, forecasting, visual recognition, bidding) powering business decisions and improving marketing efficiency by up to 50%. Earlier he co-founded a quantitative trading fund, authored the pybacktest library and built certified HFT execution infrastructure, reflecting deep expertise in performant, numerics-driven systems. Known for wearing multiple hats—product owner, architect, lead developer and mentor—he excels when innovating collaboratively with strong teams.
13 years of coding experience
6 years of employment as a software developer
Russian, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (11)
pandas10
machine-learning10
deeplearning-ai10
performance-analysis10
deep-learning10
tensorflow10
python10
implement9
nlp9
algorithm9
algorithms9
Programming languages (5)
Github contributions (5)
ematvey/pybacktest
Aug 2012 - Sep 2019
Vectorized backtesting framework in Python / pandas, designed to make your backtesting easier — compact, simple and fast
Role in this project:
Back-end Developer
Back-end DeveloperContributions:183 commits, 4 PRs, 39 pushes in 7 years 1 month
Contributions summary:Matvey completely rewrote the `EquityCurve` class and made modifications to the `EquityCalculator` class. These changes involved calculating and tracking equity changes, performance statistics, and the merging of separate curves for basket testing. Additionally, the user implemented fixes related to trade recording, merging, and the calculation of performance metrics, specifically those relating to Sharpe and Sortino ratios.
vectorizedpythoncompactbacktesting-trading-strategiestrading
ematvey/hierarchical-attention-networks
Jan 2017 - Apr 2018
Document classification with Hierarchical Attention Networks in TensorFlow. WARNING: project is currently unmaintained, issues will probably not be addressed.
Role in this project:ML Engineer
ML EngineerContributions:11 commits, 3 PRs, 23 pushes in 1 year 2 months
Contributions summary:Matvey made several changes related to model training and evaluation within the Hierarchical Attention Networks project. Their work included smoothing out evaluation mode, adjusting the evaluation frequency, and fixing vocabulary size. They also appear to have integrated with a deep text classifier, and made fixes to forward/backward concat. This suggests a focus on improving model performance and potentially refining the existing architecture.
nlpdeep-learninghierarchicaldocument-classificationmachine-learning
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.