Electrical & Computer Engineer · AI & Machine Learning · Software Engineering · Robotics
I am an Electrical and Computer Engineering graduate of Democritus University of Thrace (five-year Diploma, 2026). I build AI applications, research software and robotics prototypes. My interests span machine learning, computer vision, software engineering, autonomous systems and applied R&D.
I enjoy moving from an idea to an implementation: understanding the problem, building useful software, testing assumptions and improving reliability.
Research portfolio · Academic CV · Research projects · Contact
EYLO AI · Evidence-centered research workspace
A research software platform for discovering scholarly literature, tracing source records, exploring researchers, and organizing evidence into structured research projects. Its AI assistant, EYRA, supports analysis while keeping generated suggestions distinct from verified source material.
Visit the live EYLO website · Explore the GitHub repository
Built with React, Vite, Supabase and Vercel. Independent software project; retrieval quality and AI outputs require scientific validation.
| Project | Research focus | Evidence and status |
|---|---|---|
| SHIELD-VIO | Visual-inertial estimator health, degradation detection, and protective navigation | Research prototype; synthetic experiments and unit tests; real-world validation pending |
| DynNav | History-conditioned navigation and preservation of safe-return options under dynamic hazards | Executable planning framework and documented tests; broader statistical validation ongoing |
| Diploma thesis | Trajectory prediction of vulnerable road users at smart intersections | Academic thesis, 2026 |
My second GitHub account, PanagiotaGr, contains more projects spanning applied ML, sensing and signal processing:
- Semiconductor wafer defect classification — classical machine learning, CNNs and few-shot learning for wafer map defects.
- PC-FMCW integrated sensing, communication and illumination — predictive uncertainty for sensing-driven beam control; includes a public technical report.
- Smart intersection research literature tracker — organized literature tracking for road-user trajectory prediction.
These repositories complement my robotics projects with applied ML, signal processing and reproducible research tooling.
- Robotic perception: VIO, SLAM, multi-modal sensing, estimator consistency and failure detection
- Autonomy: uncertainty-aware decision making, online replanning and recoverability
- Machine learning: multimodal prediction, evaluation methodology and uncertainty
- Research engineering: reproducible experiments, transparent baselines and clearly reported limitations
I am interested in engineering roles, applied R&D and future doctoral research. I value reproducible experiments, maintainable software and honest reporting of what a prototype can and cannot do.
Building intelligent systems means caring about both useful functionality and reliable behavior.
Email: -