I'm Jakub Sukdol, a Computer Science Master's student at CTU FEE, specializing in Data Science and AI/ML. I bridge the gap between academic research and production-grade engineering, leveraging over 3 years of industry experience to build robust data solutions.
 |
Currently pursuing a Master's degree in Computer Science, researching Graph Neural Networks. |
 |
Specializing in Deep Learning, Scalable ML, LLMs, and Production AI pipelines. |
 |
Big fan of Open Source collaboration and Modern Terminal workflows (TUI Tools, yazi, bat, ...). |
| Timeline | Role | Company | Focus |
|---|---|---|---|
| Current | AI/ML Engineer (Contract) | Generali | Enterprise AI/ML solutions, Data Engineering, MLOps |
| Current | LLM/AI Consultant (Contract) | DevBalance | LLM integration, RAG architectures, Consulting |
| 2 Years | Software/ML Engineer | Bosch | Data Engineering, Visualization |
| Category | Technologies |
|---|---|
| Tools & Environment |      |
| Languages |     |
| Data Engineering |        |
| Visualization |    |
| ML / Data Science |       |
| ML & LLM Engineering |       |
| Software Eng |       |
| Backend (Python) |     |
| MLOps / DevOps |       |
| Misc |      |
| Project | Stack | Description |
|---|---|---|
 torch-rating |
 |
Python Package: Seamless integration of statistical rating systems into Graph Neural Networks. Full GNN implementation here. |
 PwnDépôt |
  |
Infrastructure: Secure CTF challenge hosting server with predefined tasks and CI/CD pipeline. Built with FastAPI backend, React + TailwindCSS frontend, and Dockerized environment. |
| MLP_VIZ |  |
Visualizer: Interactive visualization of MLP neural network layers and weights. Supports PyTorch models with a PyQt5 UI. |
| HyperTune |  |
Optimizer: Automatic and configurable PyTorch deep learning model optimizer developed for NPFL138 Deep Learning course. |
| General Stats | Top Languages | Streak |
|---|---|---|