Senior AI Engineer — Multimodal AI & Radiology

At Logethra, we build AI solutions for radiology. Our platform analyzes radiological images and drafts the report inside the hospital's existing workflow — the radiologist always stays in control and signs the final report. The problems below come with real data, real clinical partners, and secured resources.We are building a multidisciplinary team at the intersection of AI and healthcare, combining clinical expertise, world-class AI research, and engineering talent from institutions including ETH Zürich, METU, and the University of Groningen.We are looking for a Senior AI Engineer to work on multimodal and generative AI systems that can jointly interpret medical images and clinical text.What You’ll DoTrain, fine-tune, and improve multimodal medical foundation models that jointly process images and textResearch open-source foundation models and language models and adapt them to radiology use casesDesign architectures that bring visual and linguistic representations togetherDesign and run experiments on model architectures, training strategies, and data mixturesRigorously evaluate model accuracy, reliability, robustness, and generalizationBuild reproducible training and evaluation pipelines for systematic experimentationTrain large models using dedicated multi-GPU and high-performance computing resourcesOwn the development cycle from research experiments to prototypes and deployable software componentsWork closely with AI researchers, software engineers, radiologists, and company foundersWhat We’re Looking ForStrong hands-on development experience with Python and PyTorchStrong foundation in machine learning and deep learningExperience in at least one of:Computer VisionNatural Language ProcessingMultimodal AIExperience training foundation models or adapting pretrained models to new domainsStrong skills in experiment design, model evaluation, and systematic problem solvingSolid software engineering practices: clean, maintainable, testable code and reproducible ML pipelinesAbility to read research papers in English and translate research ideas into working implementationsSubstantial hands-on experience building and training real machine learning systemsBonus ExperienceExperience in any of the following is a plus:Medical imaging or healthcare dataVision-language or image-text foundation modelsAdapting language models to specialized domainsEfficient training and evaluation of large modelsLarge-scale datasets, distributed training, multi-GPU systems, or HPC environmentsModel reliability, explainability, and responsible AIWe do not require experience with a specific model, architecture, or training method. We care more about your ability to understand the problem, choose the right technical approach, design strong experiments, and determine what actually works.What We OfferWork on AI systems with direct clinical impact in real hospital workflowsDedicated access to large-scale European supercomputing resourcesA small, highly technical and multidisciplinary teamHigh ownership and the opportunity to influence the product and technical roadmapDirect collaboration with the foundersCompetitive salaryMeaningful equity as an early employee18 days of vacation from the first year, excluding SaturdaysFull-time, remote / flexible work within Türkiye📍 Location: Remote / flexible within Türkiye🏢 HQ: ODTÜ Teknokent, AnkaraYou do not need to tick every box. If this is the kind of problem you want to work on and your past work demonstrates strong technical ability, we would be happy to hear from you.