ML-Based User Simulation for Location Search Internship

Are you interested in how machine learning can be used to model real user behavior? Are you curious about search systems and how users interact with them, and do you enjoy experimenting with different modeling approaches? Then we would love to hear from you!

ML-Based User Simulation for Location Search Internship

What you'll do

• Design and implement user simulation agents that model realistic search behaviour.
• Explore and compare different machine learning approaches for user simulation, such as probabilistic models, supervised learning, reinforcement learning, and LLM-based agents.
• Build simulation loops that interact with TomTom’s Search APIs.
• Model user behaviours' including query reformulation, result selection, typeahead usage, and session abandonment.
• Benchmark simulated user sessions against anonymized historical data.
• Analyse search quality and behavioural metrics such as top-1 hit rate, time-to-select, session length, and reformulation rates.
• Collaborate with engineers, data scientists, and product teams to refine and validate the simulation framework.
• Contribute to reusable tooling and documentation that can be used beyond the internship.

What you'll need

• You are enrolled as a full-time MSc student for the entire duration of the internship.
• You have EU citizenship or are enrolled in a Dutch university (due to work permit regulations).
• You have a background in Computer Science, AI, Data Science, or a related field.
• You are available to start in February or March.
• Strong programming skills, preferably in Python.
• Basic knowledge of machine learning concepts and experimentation.
• Interest in user behaviour, search systems, and data-driven evaluation.
• Comfortable working with data, running experiments, and analysing results.
• Familiarity with LLM APIs (e.g. Open AI, Hugging Face) is a plus, but not required.
• A curious, analytical mindset and willingness to explore different technical approaches.