I build and evaluate AI systems for regulated and high-impact environments where performance must be measurable, outputs must be interpretable, and deployment decisions must meet operational, scientific, and governance standards.

My work spans Health AI, Urban AI, medical imaging, federated learning, multi-omics and proteomics data science, geospatial AI, and AI evaluation. I focus on turning complex data and advanced modeling into workflows that are reproducible, well-governed, and usable in real-world settings.

• Clinical AI and medical imaging

  • explainable and privacy-aware clinical AI
  • multimodal model evaluation
  • uncertainty-aware workflows
  • federated learning under regulated constraints

• Urban and geospatial AI

  • operational forecasting
  • anomaly detection
  • geospatial monitoring
  • decision support systems for public sector and urban operations

• Computational biology and life science AI

  • multi-omics analytics
  • proteomics pipelines
  • clinical genomics
  • reproducible research workflows on HPC systems

• ML systems and governance

  • MLOps
  • model evaluation
  • AI governance
  • human-in-the-loop quality control
  • stakeholder-ready technical communication

• Senior AI Engineer

  • supporting WUF13 Azerbaijan
  • supporting the State Committee on Urban Planning and Architecture of the Republic of Azerbaijan

• CTO & Co-Founder, Skolyn

  • platform architecture and R&D for explainable clinical AI
  • interoperable health data workflows
  • scalable inference services
  • privacy-aware ML systems

I have worked across research, engineering, and institutional environments including:

• Google Health
• Finnish Center for Artificial Intelligence (FCAI)
• DTU Bioengineering
• Linkoping University
• Sakerhetspolisen
• Swedish Air Force
• Bundeswehr
• European Research Council
• Nordic Energy Research
• Finnish institutions and multilateral missions to the EU, OSCE, and UN

• PhD, Systems and Molecular Biomedicine — University of Luxembourg
• MSc, Statistics and Machine Learning — Linkoping University
• BSc, Computing and Electrical Engineering — Tampere University

My academic work covers predictive modeling, biological systems, statistical learning, medical AI, computational infrastructure, and translational data science.

• Unbitrium
  Production-grade federated learning simulator and benchmarking platform for reproducible research under data heterogeneity.

• Schneileopard
  Type-safe Scala toolkit for molecular entities, omics-aware data structures, cohort modeling, pathway-aware analytics, and explainable AI workflows in systems biomedicine.

• Unbihexium
  Production-grade geospatial AI library for Earth observation, remote sensing workflows, and large-scale spatial analysis.

Health AI Medical Imaging Federated Learning Multi-omics Proteomics Bioinformatics Urban AI Geospatial AI MLOps Model Evaluation AI Governance

Languages: Python, R, Scala, SQL
ML/AI: PyTorch, TensorFlow, scikit-learn, multimodal evaluation, explainable AI
Data & Infra: Nextflow, nf-core, SLURM, HPC workflows, geospatial pipelines
Focus: reproducibility, governance, evaluation quality, maintainable ML systems

My work includes research across:

• medical AI
• explainable AI
• clinical NLP
• geospatial AI
• traffic and urban systems
• forecasting
• trustworthy evaluation
• reinforcement learning and optimization

• GitHub: @YOUR_USERNAME
• LinkedIn: YOUR_LINKEDIN_URL
• Email: olaf.laitinen@liu.se