описание
The Image and Video Understanding Lab (IVUL) and the Center of Excellence for Generative AI (CoE GenAI) at King Abdullah University of Science and Technology conduct research and work in image and video understanding and generative AI.
задачи
- Extend a multimodal agentic pipeline that interprets large, dense technical images and checks them against written specifications
- Develop exact geometric checks on vector geometry, including polygons, adjacency, enclosure, and shortest paths, and build the supporting verification layer
- Evaluate every change against expert-labelled test sets and select models based on accuracy and cost
требования
- At least four years of production Python experience, with a strong commitment to testing and measurable results
- Hands-on experience with vision-language and multimodal models, including high-resolution images, prompting, and structured outputs, and a clear understanding of their limitations
- Experience designing and building agentic workflows, including multi-step processes, tool use, and orchestration with LangGraph or similar frameworks
- Fluency in spoken and written Arabic and English
- Будет плюсом: Applied computational geometry experience using Shapely or similar libraries
условия
- One-year term, with the possibility of extension
- A technical interview and a team interview
- Applications are reviewed on a rolling basis
- Candidates must already have the legal right to work in their country of residence
