Academic depth
PhD research at AGH University on interpretable computer vision and multimodal reasoning. Published at ICCV & MICCAI. I know what’s novel, what’s reproducible, and what’s actually worth pursuing.
Open to collaborations, advisory & founding roles
I teach machines
to see -
and make it useful.
PhD computer-vision researcher and AI engineer in Kraków. I turn deep-learning research into things that ship - across sports, medicine, and trustworthy AI.
0
Papers at ICCV & MICCAI
3rd
Worldwide, DFU Grand Challenge
0
Vision domains shipped
PhD
Interpretable computer vision
PyTorch · Vision Transformers · TensorFlow · OpenCV · CUDA · ONNX · Hugging Face · Python · C++ · Docker · MLflow · Weights & Biases · Multimodal AI · Foundation Models ·
About
A researcher who loves to build.
My profile sits where academic R&D, engineering delivery, and venture creation overlap - ambitious ideas turned into prototypes, experiments, and credible products.
Engineering mindset
Comfortable across the full path - from data and experimentation to production with Python, PyTorch, Docker, and the cloud. I value systems that are robust, practical, and measurable.
Commercial instinct
I think in user pain points, proof of value, and go-to-market logic. I like technology that can leave the lab and survive contact with reality - and become a product.
Focus areas
Where I do my best work.
The domains and problem types where I contribute the most - as researcher, technical lead, advisor, or builder.
Sports analytics & movement intelligence
Turning ordinary video into coaching insight - where lightweight capture, actionable feedback, and real user value matter more than lab-perfect conditions.
Shot-technique analysis & feedback systems
Single-camera perception & performance assessment
Grassroots-friendly tools for clubs and academies
Medical AI & healthcare imaging
Image-based systems where quality, interpretability, and deployment realism matter. Strongest themes: dermatology, echocardiography, microscopy, and dental imaging.
Skin-lesion & cell analysis
Echocardiography & foundation-model workflows
Tooth & root-canal segmentation with product potential
Trustworthy & explainable AI
Systems that help people understand model behaviour, spot failure modes, and use AI in a way that earns trust - concept-based reasoning, anomaly awareness, interpretable representations.
AI commercialization & R&D strategy
Shaping the path from research result to market opportunity - clarifying the product story, finding early adopters, defining deliverables, and making the technical direction legible to partners and investors.
Selected projects
Built around real use cases.
A representative sample of the systems, products, and research directions I’ve been developing.
Single-camera football shot analysis
Automatically evaluating shooting technique from ordinary video and producing actionable feedback - combining pose estimation, vision-language models, and motion features.
Dermatology & skin imaging
Skin-lesion analysis and microscopy-driven cell understanding with segmentation networks and Vision Transformers - translating strong technical work into clinically meaningful tools.
Echocardiography AI workflows
Scalable workflows for echo-video understanding, classification, and reporting support - foundation-model-inspired approaches that improve generalization and downstream utility.
Dental CBCT segmentation
AI-assisted 3D segmentation of teeth and root canals from CBCT scans - with a product vision around dental workflows, interactive visualization, and SaaS delivery.
Research & trajectory
From interpretable vision to deployable AI.
One recurring question shapes my path: how do we build AI that’s both technically strong and genuinely useful in the world?
Interpretable computer vision & trustworthy AI
AGH University, Kraków. Concept- and structure-aware methods, spatial reasoning between high-level visual concepts, and models that don’t behave like black boxes by default.
Medical imaging & domain-specific AI
Translating advanced deep learning into healthcare - echocardiography foundation models, dermatology segmentation, dental CBCT - with emphasis on deployment realism and clinical validation.
Commercially-oriented AI for sports, medicine & industry
Building toward partnerships, demos, deployments, and startups - not only publications. Especially applied multimodal systems and human-centered decision support.
What I optimize for: technically ambitious work with a believable path to adoption.
Selected publications
Assessing the Quality of Soccer Shots from Single-Camera Video with Vision-Language Models and Motion Features
Diabetic Foot Ulcer Unsupervised Segmentation with Vision Transformers Attention
DFU-Ens: End-to-End Diabetic Foot Ulcer Segmentation with Vision-Transformer-Based Detection
Full list on Google Scholar ↗.
Work with me
Let’s build something meaningful.
I’m most useful where ambitious AI ideas need both technical depth and product realism - research collaboration, AI product strategy, and technical advisory. Based in Kraków, available internationally.