Internship: AI-assisted diagnostic and solution recommendation engine

**Start date: February 2027 · Duration: 6 months** Please include your latest academic transcripts with your application. Applications submitted without transcripts will not be considered. You will be connected to a real and active IA ELCA strategic project. You will work with real project information, collaborate closely with the ELCA project team, and, in selected situations, interact with members of the client team. As soon as the solution reaches a testable stage, it will be evaluated in an operational environment. You will join the AI Innovation team and design the AI-assisted diagnostic and AI recommendation engine. **Your tasks** * Deliver a functional prototype covering AI diagnostics and AI solution recommendation * Establish a reusable architecture for production deployment * Investigate suitable LLMs, retrieval techniques, agentic approaches and development frameworks, with support from ELCA AI architects * Design and implement knowledge ingestion, retrieval, request analysis, response generation and classification Understand the strategic, operational and business context **What we offer** * A dynamic, collaborative workplace with a highly motivated, multicultural team working across international sites * The chance to make a difference in people's lives by building innovative solutions * Internal coding events such as Hackathons and Brownbag sessions, plus our technical blog * Monthly After-Works organized at each location **About your profile** * Strong computer science fundamentals (EPFL Master's degree) and a keen interest in generative AI (LLMs, RAG) * Ability to design an end-to-end solution, from prototype to deployment; autonomous and results-driven * Good understanding of Large Language Models, generative AI and their use in real software solutions * Agility with modern development tools * Understanding of textual data processing, data analysis and information retrieval * Professional working proficiency in French and English