Andrew Muzikin
Software Engineer
Moscow, Moscow, Russian Federation
Summary
🤩
RockstarAndrew Muzikin is a software engineer with nine years' experience specializing in applied mathematics, deep reinforcement learning, and quantitative finance. Based in Moscow, he builds production-oriented research software and has applied A3C and LSTM policies to trading strategy backtesting, contributing practical examples to the btgym project that bridge Gym-style RL with Backtrader. His work emphasizes scalable, event-driven pipelines for ML-friendly backtesting, demonstrating both algorithmic depth and engineering rigor. Comfortable moving between research and implementation, he focuses on turning RL research prototypes into testable trading agents. An analytical thinker, he pairs a strong mathematical background with hands-on coding to tackle noisy, real-world financial problems.
9 years of coding experience
Stackoverflow
1reputation
904reached
0answers
1question
Github Skills (11)
lstm10
tensorflow10
python10
algorithmic-trading10
reinforcement-learning10
backtrader9
deep-reinforcement-learning9
gymnasium8
enviroment8
environ8
asynchronous6
Programming languages (4)
Github contributions (5)
Kismuz/btgym
May 2017 - Aug 2021
Scalable, event-driven, deep-learning-friendly backtesting library
Role in this project:
ML Engineer
ML EngineerContributions:619 commits, 18 PRs, 313 pushes in 4 years 3 months
Contributions summary:Andrew's contributions focus on examples related to a deep learning-friendly backtesting library. The commits show the implementation of A3C (Asynchronous Advantage Actor Critic) reinforcement learning algorithms for trading strategies using the btgym library, demonstrating the user's experience in applying machine learning models to finance. The user worked on example implementations with LSTM-based policies and contributed code related to creating, training, and testing reinforcement learning agents within the Backtrader environment, which used Breakout-v0 and Sine wave data for testing purposes.
backtestingdeep-learningevent-drivenreinforcement-learningdeep-reinforcement-learning
Kismuz/testrepo
May 2017 - Oct 2017
Contributions:32 commits, 4 PRs, 18 pushes in 5 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.