Remote AI Engineer

6 giorni fa

Milano, Provincia di Milano; Lombardia, Italia Human Technopole Tempo pieno
Build the science that shapes the future of human health. Join a place where ambitious science thrives Human Technopole (Milan) is a rapidly expanding life science institute where international researchers and advanced technologies come together to accelerate biomedical discovery and improve human health. The Ganna Group ( in the Health Data Science Centre is based at Human Technopole and the Institute for Molecular Medicine Finland, University of Helsinki. Our multidisciplinary team brings together data scientists, biologists and medical doctors to develop AI methods that integrate population-scale healthcare data with human genetics and multi-omics. We are recruiting an AI Scientist to provide methodological expertise and take ownership of AI deliverables across the group’s research programme. As AI Scientist you will develop scalable AI systems that learn from longitudinal health records and integrate genetic and molecular data to model disease trajectories, predict health outcomes and support research into personalised prevention and treatment. Your work will focus on foundation models for electronic health records, agentic AI workflows for multiverse analyses of epidemiological and biobank-scale data, and the integration of large language models with healthcare prediction models. As a senior individual contributor, you will take ownership of methodological development, from model design and training to rigorous evaluation and application, within the scientific priorities agreed with the Group Leader. Working closely with researchers and research software engineers, you will turn promising AI approaches into robust scientific tools that can be used across projects and collaborating institutions. You will also provide scientific and technical guidance to researchers, postdoctoral fellows and PhD students. You will be embedded in Human Technopole’s interdisciplinary AI environment, including the Scientific AI flagship on Multimodal AI Across Scales, and collaborate with partners at the University of Helsinki, ELLIS Institute Finland, EMBL-EBI, the Broad Institute of MIT and Harvard, and other international institutions. You will enhance your scientific and professional skills by: Developing and training foundation models for longitudinal electronic health records and population-scale healthcare data. Developing agentic AI workflows for multiverse analyses of epidemiological and biobank-scale data, automating the exploration of alternative analytical choices and assessing the robustness of scientific findings. Developing and evaluating approaches that use large language models and agentic systems to incorporate medical knowledge, interpret model outputs and support scientific analysis. Working with ICT and relevant research infrastructure teams on computing resources and secure access to sensitive data, in line with data protection requirements. Providing technical guidance and sharing AI expertise across the group and its international collaborations. Human Technopole supports career development through training, mentoring and dedicated learning opportunities. Human Technopole supports career development through training, mentoring and dedicated learning opportunities. PhD in Computer Science, Artificial Intelligence, Engineering, Mathematics, Physics, Computational Biology or related field. At least 5 years of relevant research experience, including doctoral research, in developing and applying AI methods in scientific, preferably biomedical, contexts, with a demonstrated track record of independent AI-related contributions. Strong publication record in leading AI, computational biology, or biomedical journals and conferences, demonstrating methodological innovation with real-world scientific impact. Excellent programming and software engineering skills in Python and deep learning frameworks (e.g. PyTorch, JAX). Proven experience in designing, training and applying machine learning and deep learning models, including representation learning or multimodal approaches Deep understanding of modern AI, including deep learning, representation learning, foundation models, multimodal learning, and AI agents, with the ability to translate state-of-the-art research into robust scientific tools. Strong research software engineering skills and best practices, including version control, software testing, documentation, code quality, maintainability and reproducibility, with experience developing reliable and scalable research software. Experience working with high-performance computing (HPC) and GPU environments, including the development and execution of computationally intensive AI workflows, workflow management and containerisation technologies. Experience working with large-scale scientific or biomedical datasets, particularly longitudinal healthcare records, biobanks, genetics or multi-omics , and familiarity with developing or applying foundation models and LL