Senior Applied AI Scientist, Commerce Foundation Models

Prosus (PRX) · Amsterdam

  • 2 dgn geleden

Solliciteer via de website van de werkgever

At a glance

Role
Senior Applied AI Scientist, Commerce Foundation Models
Employer
Prosus
Where
Amsterdam, Netherlands
Field
Technologie & IT
Pay
Not stated in the advert
Posted
2 dgn geleden
Sluitingsdatum
No closing date given
How to apply
Directly on the employer's own careers site

The employer's advert

Who We Are

You've ordered dinner through Just Eat Takeaway. You've bought and sold on OLX. You've paid with PayU. Chances are, you've already used a Prosus company today.

We are Prosus, a global technology company powering some of the world's leading lifestyle ecommerce businesses. Every day, billions of people across India, Latin America, Europe and beyond use our products and services to buy, sell, pay, connect and discover.

Our AI-driven ecosystem spans 100+ companies across ecommerce, food delivery, fintech and experiences, serving billions of consumers worldwide. AI powers how we operate and innovate, with more than 60,000 active AI agents helping our teams build, scale and move faster. Through our investment team, we also back the founders and technologies shaping the future: From robotics and autonomous systems to synthetic biology and frontier AI.

Guided by our five values Innovation, Results, People, Impact, and Entrepreneurship, culture is our competitive advantage. We thrive with experimentation; we build, we test, we learn, we iterate.

Join us in shaping the future of technology at global scale.

About the team

Join the AI team at Prosus, a global technology group with businesses in food delivery, e-commerce, classifieds, travel and payments. We build LCM, the Large Commerce Model: our own foundation model, trained on billions of transactions and interactions from proprietary data across the portfolio. LCM already powers search, recommendations and marketing in production at iFood. We are extending it across our companies around the world and to businesses outside the group, while broadening its capabilities in agents and real-time prediction tasks. Your work will shape what LCM learns next and feed directly into products used by millions of people.

About the role

We’re looking for experienced applied scientists in two tracks: pre-training, and post-training and continual learning. Each scientist will own a focused research direction within a track. The areas below describe the team’s research agenda; we are looking for depth in one or more of them.

The role involves designing experiments, writing the code to run them, analysing results and developing training methods that other teams can use. Scientists own the research questions, training objectives and evidence needed to decide whether a change works, and work with ML engineers to turn successful experiments into reliable training runs. We are a team of around 30 engineers with a flat structure, working directly with portfolio companies to define problems, evaluate models and bring successful work into production.

In the first three months, the expectation is to establish strong baselines in a chosen research area, develop and evaluate an improvement, and work with the relevant teams to test it in a downstream application.

Track 1: Pre-training

This track focuses on continued pre-training of open-weight models on commerce data, including how we represent that data, allocate training compute and transfer capabilities to smaller models. Research areas include:

  • Studying scaling behaviour, data mixtures and compute allocation to decide what goes into the next training run.
  • Developing mid-training methods, including data refinement, annealing schedules, context extension and alignment between text and semantic IDs for users, items and merchants.
  • Designing training sequences from user histories, merchant catalogues and market data, and studying how representations transfer across companies, languages and commerce categories.
  • Building synthetic data pipelines and measuring their contribution through controlled comparisons with real data.
  • Distilling the base model into smaller models, including through on-policy distillation, and measuring the trade-offs between quality, inference cost and latency.
  • Developing evaluations that help select pre-training runs and predict downstream performance, including checks for temporal leakage, generalisation to new users and items, and retention of existing capabilities.
  • Tabular foundation models are a possible specialisation: pre-training and in-context learning on structured commerce data, using synthetic and real tables and comparing results with established tabular methods.

Relevant experience includes continued pre-training or mid-training, data mixture research, scaling laws, tokenisation, entity representations, distillation, or tabular foundation models such as TabPFN and TabICL.

Track 2: Post-training and continual learning

This track focuses on adapting LCM to commerce tasks and learning from interactions after deployment. The goal is to turn signals from a continuous stream of billions of user sessions across the portfolio into regular updates to model weights, while retaining the capabilities the model has already learned. Research areas include:

  • Developing post-training methods using supervised fine-tuning, preference optimisation such as DPO, and reinforcement learning with verifiable or learned rewards, including GRPO and PPO.
  • Learning from session histories, searches, clicks, purchases, repeat orders and operator feedback. This includes data selection, credit assignment and reward design that account for exposure bias, delayed outcomes and the difference between observed engagement and incremental business value.
  • Studying how to combine new interactions with historical data, how frequently to update models, when an update is ready for deployment, and how to preserve performance as user behaviour and catalogues change.
  • Building RL environments with simulated users, catalogues and multi-step commerce tasks, including inspecting trajectories, validating rewards and testing whether improvements in simulation transfer to real interactions.
  • Developing reward models and LLM-based graders, calibrating them against expert judgments and observed outcomes, and testing for reward hacking, sycophancy and excessive concentration on popular items.

Solliciteer via de website van de werkgever

More openings at Prosus

Similar jobs in Netherlands

Alle secties

Editie