AI Leader · Researcher · Engineer
Gustav Olaf Yunus Laitinen-Fredriksson Lundström-Imanov
Building trustworthy AI systems that work reliably in the real world, from clinical diagnostics and bioinformatics to geospatial intelligence and strategic technology leadership. Currently completing dual PhDs at the Technical University of Denmark and University of Luxembourg.
11+
Publications
25+
Roles & Engagements
8
Skill Domains
5+
Countries
Career
Featured Experience
Machine Learning Engineer
Säkerhetspolisen (SÄPO)
Builds and operates AI applications that make machine learning and data science reliably usable across the agency. Works end-to-end across problem framing, data readiness, model training, evaluation, deployment, and lifecycle management, with strong emphasis on reliability, traceability, and responsible AI in a security-critical setting. Contributes to shared data platform development, reproducible pipelines, automated testing, CI/CD alignment, monitoring, and production-grade LLM and multimodal model deployment. Operational details remain confidential.
Data Science Specialist in Proteomics
Technical University of Denmark, Department of Bioengineering
Designs, implements, and optimizes large-scale computational workflows for mass spectrometry-based proteomics. Leads development of five bioinformatics pipelines in Python and R processing over 10 TB of raw proteomic data annually, spanning peptide identification (MaxQuant), normalization, differential abundance analysis (limma, DEqMS), and functional enrichment (GO/KEGG). Optimization work reduced pipeline runtime by 30% while improving reproducibility across six SLURM HPC environments. Implemented Nextflow and nf-core standards. Developed an integrative ML framework for early-stage neurodegeneration biomarkers, improving AUC from 0.87 to 0.94 across validation cohorts exceeding 2,000 samples.
Chief Technology Officer & Co-Founder
Skolyn
Leads end-to-end technology strategy and R&D for an explainable clinical AI platform designed to reduce diagnostic error in radiology and pathology. Architected the Skolyn AI Platform, integrating HL7/FHIR-based data pipelines, enterprise APIs, and distributed inference capable of analysing over 127 pathological indicators per scan in under 3 seconds. Oversaw a modular deep learning stack spanning X-ray, CT, and MRI with >95% diagnostic accuracy and visual interpretability. Leads ML engineers, backend developers, and clinical data scientists. Guides product design, model optimization, regulatory readiness, privacy-preserving training, and enterprise SaaS commercialization. Scaled infrastructure to support over 50,000 scans during pilot programmes and contributes to fundraising strategy.
Research
Recent Publications
Spatiotemporal Heterogeneity of AI-Driven Traffic Flow Patterns and Land Use Interaction: A GeoAI-Based Analysis of Multimodal Urban Mobility
Urban Spatio-Temporal Foundation Models for Climate-Resilient Housing: Scaling Diffusion Transformers for Disaster Risk Prediction
Autonomous AI Agents for Real-Time Affordable Housing Site Selection: Multi-Objective Reinforcement Learning Under Regulatory Constraints
Open to research collaborations and advisory engagements
Interested in medical AI, responsible systems, translational research, or technical strategy? Let's connect.