<div class="show-more-less-html__markup show-more-less-html__markup--clamp-after-5 relative overflow-hidden"> <p><strong>About the role</strong></p><p><span>We are looking for an AI/ML Engineer to build production AI systems that power </span>PropX's Digital Twin platform, specifically designed for managing entire building portfolios and automating ESG intelligence. <span>You will develop computer vision models, knowledge graph–based retrieval systems, and LLM-powered applications that transform geospatial, BIM, and infrastructure data into actionable intelligence.</span></p><p><br/></p><p><strong>What you'll do</strong></p><ul><li><span>Fine-tune computer vision models </span></li><li><span>Build a RAG pipeline on a graph database</span></li><li><span>Run a local small language model and escalate to a larger model when needed</span></li><li><span>Set up model versioning, scheduled retraining, and rollback, so regressions get caught before they ship</span></li><li><span>Build APIs that hand model outputs, with confidence and source attached, to the rest of the backend</span></li></ul><p><br/></p><p><strong>Minimum qualifications</strong></p><ul><li><span>Master's in AI / ML or related fields</span></li><li><span>Programming experience specifically in Python, APIs, SQL, Git, testing, Docker.</span></li><li><span>Experience fine-tuning a pretrained computer vision model (detection or segmentation) on a real dataset</span></li><li><span>Experience building a RAG pipeline: retrieval, prompting, grounding output in retrieved data</span></li><li><span>Experience running an open-weight LLM (Mistral, Llama, Phi-3, or similar), not just calling a hosted API</span></li><li><span>Experience versioning, monitoring, or rolling back a model in production</span></li><li><span>Experience designing an API that other services depend on</span></li></ul><p><br/></p><p><strong>Preferred qualifications</strong></p><ul><li><span>Experience with satellite or street-level imagery</span></li><li><span>Graph-based RAG using Neo4j / knowledge graph / vector search.</span></li><li><span>Experience with a cloud platform (AWS, GCP, or Azure) and Docker in a live deployment</span></li></ul><p><br/></p><p><span>Submit your application by </span><strong>20 July 2026</strong><span>. Applications will be reviewed on a rolling basis, so early applications are encouraged.</span></p><p><br/></p><p><br/></p> </div>