Aristotelis Tsoutsanis
AI/ML Engineer & Researcher
aristotelis@tsoutsanis.com · linkedin.com/in/aristotelis-ts · github.com/Aristotelis1 · tsoutsanis.com
Profile
AI/ML Engineer with experience across Computer Vision and NLP. Proven track record in Vision-Language Models, autonomous systems, and traditional ML.
Experience
Machine Learning Engineer — OSeven Telematics
- Developing machine learning models to improve road safety and driving behavior through telematics.
Machine Learning Engineer — PreciTaste
- Pioneered research on Vision-Language Models (VLMs) with resource-efficient architectures using pre-trained backbones.
- Enhanced deep learning pipelines for reliability and speed.
- Developed advanced backend services, including an innovative auto-labeling tool powered by Segment Anything.
Education
PhD, Machine Learning and Road Safety — National Technical University of Athens
- Conducting research in road safety using telematics data.
- Implemented a framework that detects harsh cornering events using mobile sensor data from drivers.
- Supported by the IVORY Project, funded by Marie Skłodowska-Curie Actions (MSCA).
Master's in Computer Science — Technical University of Munich
- Specialized in Computer Graphics, Machine Learning, and Robotics.
- Thesis and project work on the Smartfield yield forecast ensemble and vision-based autonomous driving (below).
Bachelor's in Computer Science — National and Kapodistrian University of Athens
- Completed advanced projects in machine learning.
- Developed image classification models and predictive systems.
Selected Projects
Smartfield Yield Forecast Ensemble
Built ensemble models for precise fertilizer and growth regulator scheduling, boosting yields via data-driven interventions. Forecast deviations under 2 days. smartfield.ai
Vision-based Autonomous Driving
End-to-end lane keeping and steering control prediction; experiments on Udacity and TU Munich simulators and datasets. Udacity repo · TUM repo
Technical Skills
- Languages: Python, C/C++
- ML & Vision: PyTorch, OpenCV, Segment Anything, Vision-Language Models
- Data: NumPy, Pandas, PySpark
- Infra & Tooling: MLflow, FastAPI, Git, Google Cloud, AWS