Data Scientist Analyst

About AXAAs a world-leading insurance company, we act for human progress by protecting what matters. With 153,000 employees in 54 countries working for 105 million customers, we’ve created a truly dynamic and vibrant community. Inclusion and diversity link closely with our values, and together we’re nurturing a culture of respect, for each other, for our customers and the communities around us. Join AXA and you’ll feel like you belong, are included and can thrive. You’ll be able to shape the way you work and truly grow your potential as you seek out new opportunities, push boundaries and benefit people in critical moments of their lives. This is your chance to build the tomorrow you want. Know you can.About The EntityAXA is becoming a sustainable tech-led company and at AXA Group Operations we are one of the major catalysts for this transformation.We set the tone by triggering and empowering the evolution of our insurance business model through technology and innovation, driving its concrete implementation globally at speed, with a high quality of advisory and execution.We are present across 17 countries with committed, highly qualified teams. We leverage technology, data, sourcing, security and investment allocation in a global way, but also achieve economies of scale and synergies when necessary.At AXA Group Operations, we want to be recognized in three fields of action:State-of-the-art Data Technology to drive customer experienceState-of-the-art Procurement & Sourcing to drive efficiency and better manage risksHigh-Performing Global Team for stronger partnerships with AXA entities Job position pitch Where will you be in the organization?The department / team Within AXA Group Operations, Group Data, AI & Innovation (GDAI) explores and scales the value of data, AI, digital transformation, emerging technologies (IoT, Geospatial, Blockchain, Quantum Computing, etc.) as well as innovative business ideas with the potential to disrupt the current insurance business model and to shape future opportunities to better partner in our customers’ lives. GDAI and its spin-offs consist of 200+ inspiring minds located in Paris, Barcelona, London, Lausanne (on the EPFL campus) in Lausanne, and on San Francisco (on the Stanford campus) incubating and deploying solutions in co-creation and collaboration with AXA operating entities and strategic partners around the world.About The JobJob purpose Current Vision-Language Models (VLMs) demonstrate impressive capabilities on natural images but struggle with insurance-specific aerial imagery analysis. Generic models can identify broad categories like "airplane" or "vehicle" but fail at fine-grained distinctions critical for commercial insurance risk assessment—such as differentiating wood pallets from containers, identifying specific aircraft models, or detecting dust collectors and cooling towers. Traditional approaches require 1-3 months of dedicated data science effort per object type, making comprehensive portfolio analysis impractical.This project aims to break through these limitations by developing specialized VLMs capable of understanding insurance-relevant objects and spatial relationships in aerial imagery through natural language queries, enabling risk engineers to efficiently analyze entire portfolios without building detectors for every possible object type.Main missions Develop And Validate Fine-grained Vision-language Models For Aerial Imagery That Enable Natural Language-based Retrieval And Classification Of Insurance-specific Objects. Key Deliverables IncludeConstructing fine-grained remote sensing caption datasets using VLMs, validated through human preference studiesImplementing multimodal structured embeddings using graph-based and scene-based representationsDeveloping novel retrieval strategies (Graph-to-Graph, VectorGraph-to-VectorGraph) with Graph Matching NetworksCreating lightweight scene-based matching approaches for resource-constrained environmentsDemonstrating significant performance improvements over baseline CLIP-based methods on insurance-relevant benchmarksExpected Skills & ExperienceWe are looking for someone with the following experience and skills:ExperienceMaster's degree student or recent graduate in Computer Science, Data Science, or related fieldDemonstrated research experience in computer vision, natural language processing, or multimodal learningPrior work on remote sensing applications, vision-language models (CLIP, BLIP, etc.), or graph neural networks highly valuedExperience with academic writing and publishing preferredRequiredTechnical skillsStrong programming proficiency in Python and deep learning frameworks (PyTorch/TensorFlow)Experience with vision-language models and transformersFamiliarity with computer vision fundamentals (object detection, segmentation, image retrieval)Knowledge of contrastive learning and multimodal embeddingsPreferredExperience with large language models (GPT-4, Gemini) and prompt engineeringUnderstanding of graph neural networks and scene graph generationFamiliarity with remote sensing imagery and geospatial dataKnowledge of retrieval metrics and evaluation methodologiesExperience with SAM (Segment Anything Model) or similar segmentation toolsSoft Skills / Transversal SkillsResearch autonomy: Ability to work independently on complex problems while maintaining regular communicationAnalytical thinking: Strong problem-solving skills with attention to fine-grained detailsCollaborative mindset: Experience working in cross-functional teams with domain expertsAdaptability: Comfortable pivoting between different approaches (graph-based, scene-based, hybrid methods)Communication: Clear documentation and presentation skills for technical and non-technical stakeholdersPragmatism: Balance between research innovation and practical, deployable solutions for insurance applicationsDuration: 3 months (40% part-time)What We OfferWe bring together the expertise, cultural diversity and creativity of over 8,000 employees worldwide and we’re committed to equal opportunities in all aspects of employment (gender, LGBT+, disabled persons, or people of different origins) and to promoting Diversity & Inclusion by creating a work environment where all employees are treated with dignity and respect, and where individual differences are valued.