About the role
Generative AI Engineer at Leonardo, inside e-GEOS (the Telespazio and ASI joint venture) in the Space Division, based in Rome. The team builds next-generation Earth Observation applications, combining LLMs, RAG, Geospatial Foundation Models and agent-based workflows so users can query satellite imagery, technical documentation and environmental data in natural language.
You would design and ship production AI applications rather than do foundational model research: retrieval systems, AI copilots, agentic pipelines, knowledge graphs of EO entities, and evaluation frameworks for grounding, hallucination and cost. The stack leans on LLMs, RAG, embeddings, semantic and hybrid search, and Python. It is pitched as a junior role with a clear path toward owning the Earth Observation GenAI architecture, mentored by the technical lead.
Highlights
- Design and ship production GenAI: retrieval systems, AI copilots, agentic pipelines, knowledge graphs of EO entities
- Stack: LLMs, RAG, embeddings, semantic and hybrid search, Python
- e-GEOS (Telespazio and ASI joint venture), Space Division, based in Rome
- Junior role with a path toward owning the Earth Observation GenAI architecture; modern GenAI applied to satellite intelligence at one of Italy's largest aerospace and defence groups
This recap is dataskew's editorial summary, not the company's copy.