(MSc/PhD) AI Research Intern - LLMs, Causal Inference & Decision Making

Prosus; Amsterdam; On-site; EUR 13,000 - 23,000; Full time

Benefits offered by this job

Learn from the best

Hard ML problems

Unique data at scale

Real infrastructure

Ship real work

Grow your career

Junior Engineer compensation

Best-in-class tools

Job summary

Prosus in Amsterdam invites you to join a driven AI team working on applying LLMs and modern machine learning to causal decision-making at scale. You will build models, run experiments, and transform research ideas into real systems while leveraging vast user data and leading ML tools.

The internship emphasizes mathematics, statistics, and quantitative reasoning, with opportunities to explore uplift modeling, treatment effects, and constrained optimization in real-world promotions and incentives.

Qualifications

• Strong foundation in mathematics, probability and statistics.
• Solid quantitative modeling fundamentals.
• Experience with causal and experimental reasoning.
• Interest in optimization and decision making.
• Proficient Python programming.

Responsibilities

• Build and experiment with LLMs, representation models, and modern ML systems for causal problems.
• Work with large-scale data to study user responses to interventions.
• Develop models for treatment effects, uplift, and counterfactual prediction.
• Design experiments, analyze randomized controlled trials, and evaluate results.
• Develop constrained optimization models for promotions and incentives.
• Read papers, prototype ideas, and turn them into production systems.

Skills

Math & stats

Quant modeling

Causal reasoning

Optimization

Python

ML fundamentals

PyTorch/JAX

Research literacy

Teamwork

AI assistants

Tools

PyTorch/JAX

Job description

Join our AI team at Prosus, the largest consumer internet company in Europe and one of the biggest tech investors in the world. You’ll work on applying LLMs and modern machine learning to causal decision-making problems at large scale, including promotion and incentive systems.

What you’ll do:

• Build and experiment with LLMs, representation models, and modern ML systems for causal and decision-making problems

• Work with large-scale experimental data to understand how users respond to different interventions

• Build models for treatment effects, uplift, and counterfactual prediction

• Design experiments, analyze randomized controlled trials, and evaluate results

• Develop constrained optimization models for promotion and incentive allocation

• Read research papers, prototype ideas, and turn the ones that work into real systems

Preferred qualifications & traits:

• You’re strong in mathematics, probability, and statistics and enjoy reasoning about problems formally.

• You have strong quantitative modeling fundamentals , whether your background is in machine learning, economics, finance, operations research, statistics, or another technical field.

• You understand causal and experimental reasoning and are comfortable thinking about RCTs, counterfactuals, treatment effects, and confounding.

• You’re interested in optimization and decision making , such as allocating limited resources, optimizing under constraints, or choosing actions under uncertainty.

• You know how to write good Python code from projects, research, coursework, or similar experience.

• You’re comfortable with modern AI , including neural networks, representation learning, transformers, and LLMs, or have the technical background and interest to learn them quickly. Experience with PyTorch/JAX is a strong plus.

• You can make sense of research by reading papers, understanding the key ideas, questioning assumptions, and turning them into working code.

• You work well with others , are genuinely curious, and enjoy learning new ideas and technologies.

• You’re friends with AI assistants and use them regularly for coding, research, and writing.

• You can commit to 6-12 months working with us in Amsterdam (minimum 3 days in the office weekly). We’ll work with your academic schedule.

Experience with causal inference, econometrics, empirical economics, operations research, quantitative finance, uplift modeling, treatment-effect estimation, or mathematical optimization is a plus. This could include methods such as S/T/X-learners, causal forests, causal DAGs, potential outcomes, linear or integer optimization, Lagrangian/dual methods, stochastic optimization, or related techniques.

Prior experience with promotions, pricing, ads, recommendations, logistics, marketplaces, or other allocation and decision systems is a strong bonus, but not required.

What we offer:

• Learn from the best : Work with AI engineers who’ve published at NeurIPS, released top models on Hugging Face, and built production systems at scale

• Work on hard ML problems : Combine LLMs, causal inference, experimentation, and optimization in real-world decision systems

• Unique data at scale : Access to billions of interaction logs from 500+ million users across our e-commerce platforms

• Real infrastructure : Access to H200 GPU clusters, large-scale datasets, and all major LLM APIs (OpenAI, Anthropic, Google)

• Ship real work : Prototype ideas, test them rigorously, and work toward systems that influence real product decisions

• Grow your career : People who excel here tend to stick around

• Junior Engineer compensation : We pay competitively for the impact you’ll have

• Best-in-class tools : MacBook Pro, access to major LLM APIs and premium coding assistants

• Great workspace : Modern office in Amsterdam South with free barista coffee and NS travel card for your commute

This internship puts you at the intersection of modern AI, causal inference, and real-world decision making. If you like mathematics, experimentation, machine learning, and LLMs, and want to work on problems where models are used to make real decisions, we’d love to hear from you.

Our Diversity and Inclusion Commitment

We respect the dignity and human rights of individuals and communities wherever we operate in the world. Building an inclusive workplace where everyone feels welcome and can thrive is critical for us. We provide access to education, which helps everyone understand the important role they play and the positive impact they can have.