Rokas Liuberskis is a Senior Software Engineer and machine learning specialist with 7 years of experience building computer vision and reinforcement learning systems for production and research. He founded PyLessons to teach Python and ML topics and has hands-on experience implementing YOLOv3 in TensorFlow and developing RL trading agents, reflecting a blend of practical engineering and research curiosity. Rokas has delivered ML-driven solutions across industries—from face verification and document liveness at Bankera to routing and driver-load recommendation systems in logistics—while maintaining robust data pipelines and real-time preprocessing logic. Comfortable moving from low-level electronics and SCADA automation to high-level model design, he thrives on tinkering and iterating until systems are both elegant and reliable. Based in Klaipėda, Lithuania, he pairs a master’s in information systems engineering with a maker’s mindset and a habit of improving his craft every day.
7 years of coding experience
10 years of employment as a software developer
Bachelor's degree, Electrical and Electronics Engineering, Bachelor's degree, Electrical and Electronics Engineering at Klaipeda University
Master's degree, information system engineering, Master's degree, information system engineering at VILNIUS TECH - Vilnius Gediminas Technical University
Trying to create Reinforcement Learning powered Bitcoin trading bot
Role in this project:
ML Engineer
Contributions:34 commits, 1 PR, 35 pushes in 1 year 10 months
Contributions summary:Rokas contributed multiple versions of a Python script designed to build a Bitcoin trading bot using reinforcement learning. The code includes environment setup, action spaces, reward calculations, and a custom environment class. The user progressed through several iterations, adding visualization tools, and implementing a shared model architecture. The later commits also incorporated indicators and multiprocessing to enhance the bot's functionality and training efficiency.
Contributions:215 commits, 9 PRs, 214 pushes in 2 years 5 months
Contributions summary:Rokas contributed to the YOLOv3 implementation in TensorFlow, focusing on the core functionality of the model. They modified the `utils.py` file, adding and updating functions related to image preprocessing, bounding box handling, and non-maximum suppression (NMS). The user was likely involved in integrating and refining the core object detection pipeline within the TensorFlow 2.x framework, specifically for YOLOv3.
tensorflowyoloyolov4trackingdetection
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.