MSc Thesis: Agentic AI for Configuring Radiotherapy Workflows

RaySearch develops innovative software solutions to improve cancer care. Over 1200 clinics in more than 54 countries use RaySearch software to improve treatments and quality of life for patients. RaySearch was founded in 2000 and is listed on Nasdaq Stockholm. The headquarters is located in Stockholm, with subsidiaries in the US, Europe, Asia and Australia & New Zealand. Today we are more than 485 employees with a common vision of improving cancer care with innovative software. Our great staff is crucial for our success and we offer a fantastic working environment in modern offices, flexibility and good opportunities for development. We believe in equal opportunities, value diversity and work actively to prevent discrimination. Are you a problem-solver with a passion for mathematics and optimization? Join us in exploring new ways to improve adaptive radiotherapy and cancer treatment. Project description This project develops AI agents that take a description of how a radiotherapy clinic works and build the matching workflows in the RayCare oncology information system. RayCare is developed by RaySearch at our headquarters in Stockholm, where the thesis work will take place. In RayCare, each clinic's way of working is configured as workflows: tasks for different staff roles, grouped into modules and linked by triggers that decide when each task becomes ready and when it is due. Today these are built template by template through forms, which takes RaySearch staff and clinicians substantial time and raises the barrier to entry for new clinics. Reducing this effort will make it easier for RayCare to scale as it enters clinical use at a growing number of clinics around the world. The project will develop and evaluate the most effective approach to a full round trip: from a clinic's description of its process to a configured workflow, and back to a representation that a human can review. The workflow the agents produce must then be correct, and the project will establish how to measure this: a workflow can look plausible yet contain a missing dependency or a task assigned to the wrong role, while two workflows can differ structurally and still behave identically. Historical data from clinics using RayCare makes this measurable: the original specifications can serve as input to the agents, and the workflows configured from them, available in structured form, as ground truth. You will build the agents with the GitHub Copilot SDK, with full access to the latest frontier AI models, and integrate them with RayCare. Your profile Requirements:

• Highly motivated students that have excelled in their studies in an engineering program with a focus on algorithms and programming, such as computer science, machine learning, electrical engineering, or engineering physics
• Knowledge and interest in programming.
• Excellent communication skills in Swedish and/or English

Meriting:

• Experience gained from summer/part-time jobs or hobby projects within software development and/or machine learning
• Experience from working with agentic AI e.g. Claude, Codex, GitHub Copilot, etc.
• Programming language skills such as C#, TypeScript

Application Please apply to the position through the link below. Selection and interviews will be ongoing. We do not accept applications by e-mail. Application Link.