AI & Computer Vision Consultant

20+ years turning pixels into decisions.

From a medical imaging PhD, to real-time drone detection, to leading digital pathology AI at scale — now working independently with teams who need senior computer vision and AI expertise.

  • PhD in Medical Image Analysis
  • Ex-Director of AI Engineering (Ultivue / Vizgen)
  • Former Hensoldt, Definiens
  • 25+ scientific publications
Portrait of Ruben Cardenes

Services

What I can help you build

Senior, hands-on expertise across the full computer vision and AI lifecycle — from a single hard problem to a full team.

Computer Vision Engineering

Object detection, segmentation, co-registration, classification and tracking — designed, trained and shipped to production.

DETR D-FINE YOLO U-Net Vision Transformers Optical Flow Siamese Networks

AI & Computer Vision Consultancy

Solution architecture, technical feasibility assessments, MLOps strategy, technical due diligence for investors/M&A, and team training.

Solution Architecture Feasibility Studies MLOps Technical Due Diligence Team Training

Agentic AI

Design and implementation of AI agent systems, workflow automation, and integration with large language models.

Agent Design Workflow Automation LLM Integration Tool Use

Domains of Application

Where this expertise has been proven

Deep, applied experience in demanding real-world domains — not just benchmarks.

Microscopy & Digital Pathology

mIF, IHC and H&E image analysis — 5 years leading AI for digital pathology at Ultivue / Vizgen.

Real-Time Video Processing

Sports analytics, drone detection, people tracking and embedded deployment on Jetson — built at Hensoldt.

Medical Imaging Research

MRI, CT, DT-MRI and vascular modeling — rooted in a decade of academic and postdoctoral research.

Featured Projects

From GitHub

A selection of open work — the rest lives on GitHub.

Raspberry Pi Autonomous Car

A self-driving toy car built on Raspberry Pi, using on-board computer vision for lane and obstacle detection. Includes demo video.

Python OpenCV Raspberry Pi

Geodesic Distance Transform

A C implementation of the geodesic distance transform, used for image segmentation and shape analysis research.

C Image Processing

TODO: Project name

TODO: one-line description of the project.

TODO

TODO: Project name

TODO: one-line description of the project.

TODO

TODO: Project name

TODO: one-line description of the project.

TODO

TODO: Project name

TODO: one-line description of the project.

TODO
20+
Years of experience
27
Publications (19 conf. + 8 journal, 1st author)
14
R&D projects delivered
7
People led in a single team

About

From physics to AI, across three countries

My path started in physics, followed by a PhD in medical image analysis, and postdoctoral research at Harvard and the BESSY synchrotron. From there I moved into industry — Definiens, Hensoldt, and Ultivue/Vizgen — building computer vision systems for medical imaging, defense and digital pathology at increasing scale and responsibility, eventually leading a team of 7 as Director of AI Engineering.

Today I work independently, partnering with companies and teams who need senior, hands-on computer vision and AI expertise without the overhead of a full-time hire.

Location Munich, Germany — remote / hybrid
Education PhD, Medical Image Analysis
Research Postdocs at Harvard & BESSY synchrotron
Industry Definiens · Hensoldt · Ultivue / Vizgen
Languages Spanish (native) · English · German (C1)

Contact

Let's talk about your project

Tell me a bit about what you're working on — I'll get back to you within 1-2 business days.

Location

Munich, Germany — available remote / hybrid