Hi, I'm Reza Karbasi.
A
Welcome to my professional portfolio. I'm Reza Karbasi, a Machine Learning Specialist and Electrical Engineer with a deep passion for solving complex, real-world problems through the power of AI and robotics. With a solid foundation in electrical engineering from Khaje Nasir University of Technology and advanced expertise in AI and robotics from Tehran University, I've dedicated my career to developing innovative solutions in machine learning, from advanced driver assistance systems to speech recognition technologies. Explore my journey and discover how I leverage technology to push the boundaries of what's possible.
Link to my PDF resume!
About
From an early fascination with robotics to a profound career in artificial intelligence, my academic and professional journey is driven by a quest for continuous learning and improvement. Holding a Bachelor's degree in Electrical Engineering and a Master's in AI and Robotics, I've not only acquired theoretical knowledge but have also applied these concepts in practical, impactful projects. My work spans developing path planning algorithms for autonomous vehicles at SoftwareMotion to enhancing speech recognition models for real-world applications. Each project has been a step towards my goal: to innovate and contribute to the advancement of intelligent systems.
- Highlight: Machine Learning, Reinforcement Learning, Machine Vision, Speech Recognition
Experience
- Developed and deployed a face liveness detection system, reducing FRR by 10% in production.
- Implemented a thermal imaging system for solar panel fault detection using YOLOv8, deployed on a 12.7 MW solar farm.
- Created a car tracking system for surveillance cameras, fine-tuned models for lane and vehicle identification.
- Enhanced OCR system with EasyOCR, Pytest for testing, and MLflow for experiment tracking, improving accuracy.
- Topics: Face Anti-Spoofing, Object Detection, Computer Vision, OCR, Surveillance, Deep Learning, Machine Learning
- Tools: Python, PyTorch, TensorFlow, OpenCV, YOLOv8, MLflow, Docker, Git, FLIR Thermal Cameras
- Spearheaded codebase analysis for ADAS levels 2+ and 3, enhancing automotive safety and efficiency.
- Innovated a dynamic programming-based path planner, significantly improving route optimization.
- Developed a visualization app for AEB and ACC functionalities, streamlining system analysis and debugging.
- Deep analysis of STT models.
- Learning Audio-classsification models
- Employ Keyword spotting methods to accomplish some tasks.
- Topics: Speech-To-Text, Audio-Classification, Signal-Processing, Deep-Learning, Machine-Learning
- What is RL
- Markov Decision Process
- Dynamic Programming
- Temporal Difference methods
- Deep RL methods : A3C, DDPG
- Designed a reinforcement learning agent to automate trading strategies, achieving a significant improvement in trading performance.
- Published findings and methodologies on my Medium blog, contributing to the broader knowledge base in AI-driven trading. Link to the blog
- Internal Communication Device for a hospital : board design, microcontroller programming
- Taxi Paying Device: making a simple device to communicate with driver while transfering passengers. ESP32 programming, Mechanical design and board design
Featured Publications and Blogs

RL agent trader using mql5 in metatrader platform for automated trading.
Skills
Languages






Libraries





Frameworks






Other




Education
Tehran, Iran
Degree: Master of Science in Artificial Intelligence and Robotics
GPA: 18.83/20
- Reinforcement Learning
- Deep Learning
- Machine Vision
- Advanced Robotics
- Artificial Intelligence
- System Identification
- Big Data
- Social Network
- Data Analytics
Courses:
K. N. Toosi University of Technology
Tehran, Iran
Degree: Bachelor of Technology in Electrical Engineering
GPA: 17.4/20
- Pattern Recognition
- Digital Control
- Signals and systems
- Modern Control
- Advanced Programming
- Statistics and Probability
- Linear Algebra
Significant Courses: