Thesis: AI-Powered Recommendations in Medical Search

Hello, we are AMBOSS and looking for a Working Student to join our Product Team!

About AMBOSS

AMBOSS is a learning and clinical decision support tool striving to empower physicians across the globe to provide the best possible care. Our founders set out in 2011 to create a tool that they would have hoped to have as medical students and doctors. Since then we have grown to currently operate in 180 countries and have gained immense traction in Germany and the US. Currently, we are pursuing this mission with more than 500+ employees in our offices in Berlin, Cologne, New York, and Cagliari.

Why can this position be exciting for you?

At AMBOSS, we’re redefining how medical knowledge is accessed. As a Working Student, you’ll work on a thesis project that will enhance our search platform by improving semantic retrieval, recommendation and ranking algorithms. You’ll support our Product Manager and collaborate closely with senior AI Specialists and Developers, so your work will directly impact medical education and healthcare by helping users find high quality content efficiently. This is a unique chance to advance cutting-edge AI technology while making a meaningful global impact.

What you will do:

• Work on optimizing the recommendation and ranking algorithms for semantic retrieval, focusing on broad medical topics.
• Conduct literature research to learn different recommendation algorithms and how they can be used in our hybrid retrieval pipeline (BM25 and Dense Retrieval using Embeddings).
• Collaborate closely with our Search Team, including Product managers, Software developers, and AI Specialists, to understand the challenges of semantic retrieval.
• Define and implement an evaluation strategy for different approaches to enhance user experience by making broad searches more relevant and insightful.
• Contribute to building and testing models, using data-driven methods to evaluate the impact of algorithm improvements.
• Implement an Online Randomised Controlled Trial (RCT, A/B test) measuring the impact of the improved recommendations for our users.
• Present findings and improvements to stakeholders, providing insights into how these changes enhance the user journey.

What you will bring: 

• A background in computer science, machine learning, data science, or related fields.
• You're familiar with information retrieval, semantic search, or recommendation systems.
• Strong interpersonal skills to work closely with the search team, including AI specialists, developers, and product managers.
  

You enjoy

• Python programming with common data science libraries.
• Evaluating AI systems using offline and online experiments.
• Eagerness to learn, iterate, and adapt in a dynamic environment.
• You enjoy tackling challenging problems, especially those that require creative solutions in areas like semantic retrieval and machine learning.
• You find purpose in knowing that your work directly contributes to improving medical education and, ultimately, patient care.
• You thrive in an environment where experimenting with new approaches, learning from data, and iterating on your findings is part of the process.
• You enjoy working with a diverse team, bridging the gap between technical research and product-focused development.
• You have a drive to continuously enhance algorithms and user experiences, always seeking ways to deliver more relevant and precise information to users.

Benefits

AMBOSSians tell us that innovative work keeps them energized and employee benefits help them to feel appreciated and empowered. We invest in every AMBOSSian with our employee benefits package, crafted to support financial, physical, and mental health, and work-life harmony.

Check out all of our employee benefits below: 

https://go.amboss.com/the-amboss-prescription-de

We believe in diversity as a driving force of innovation and welcome people of all backgrounds to help us achieve our mission of empowering physicians to provide the best possible care – to everyone, everywhere.

Did we just describe your ideal next role? We encourage you to apply even if you do not meet all of the requirements.