Product Data Science & AI Agents Intern

Criteo is looking for new talented interns able to start from January 2027!

The Product Analytics & Data Science team (PAX) uses cutting-edge technology, advanced statistics, machine learning and — increasingly — AI agents to tackle some of the most complex Product challenges at Criteo. We are evolving from a classic Data Science function toward an AI-native team: embedded alongside Product, R&D, GTM and Design, we frame the right problems early, prototype fast, and build trusted, reusable AI systems that help Criteo ship better products, faster, harnessing Product Intelligence.

Think "diagnose the messy problem and build the tool that solves it": our approach combines product judgment, data, and hands-on technical fluency.

In this internship, building AI agents and skills is at the center of the role. You will design, prototype and evaluate agents that change how PAX and our Product partners work. Agents you build and validate can graduate into owned, production capabilities for the team.

Wondering how life is in the PAX team? Take a peek at Live my life with the Product Analytics & Data Science team.

You will be assigned to one or several projects. The topics we tackle are wide and always evolving. Recent examples of the AI-native, agentic work you could contribute to:

• Build AI agents & skills for the team — design agents that automate recurring analytical work (e.g. a reporting agent, a Slack assistant, an AI skill), and ship them into our shared skills registry so the whole team benefits.

• Agent evaluation & guardrails — help build frameworks that evaluate multi-turn agents (LLM-driven personas talk to a real agent, LLM judges score relevance, coherence, helpfulness, safety and accuracy), so agentic products ship faster and more safely.

• Product intelligence prototypes — build retrieval / embedding-based systems for product understanding (e.g. semantic search over publisher content for supply discovery, brand safety, or audience creation).

• LLM-powered diagnostics at scale — use LLMs to detect, sample and route quality issues across complex product pipelines (e.g. tracing where a catalog signal is lost and who should fix it).

• Messy, cross-functional product problems — jump into ambiguous product and client challenges (identity, measurement, trading, creative validation) and use fast prototyping and AI tooling to converge on solutions that work in the real world.

Overall, your responsibilities include:

• Design, build and evaluate AI agents and reusable skills that solve real product and analytical problems

• Support the team in framing product and client problems early with PM, R&D, GTM and Design — then prototype narrow, high-value solutions quickly

• Build evaluators, guardrails and golden-dataset tests so the systems you ship are trustworthy

• Turn repetitive manual tasks into trusted agents that scale beyond you

• Mine large data sets and turn them into understandable and actionable insights

• Master our internal analytic datasets, tooling and AI stack

• Master's degree student or higher (Mathematics, Computer Science, Physics, Engineering, Economics, etc.)

• Available for at least 5 to 6 months from January 2027.

• End of study or gap year internship

• Outstanding analytical skills, creative thinking, and a builder's mindset — you like to prototype, not just analyze

• Fluency in the core toolkit of Data Science: Python; SQL; manipulating large-scale data sets; building data pipelines; descriptive and predictive modeling; implementing visualizations, dashboards and reports

• Curiosity and ideally hands-on experience (through projects, coursework or personal tinkering) with the AI-native toolkit: LLMs and prompting, building agents or skills, retrieval / embeddings, or agent evaluation — or a strong eagerness to learn these fast

• Excellent interpersonal and communication skills, pro-active and independent to work with!

PS: Please apply with an English resume

We acknowledge that many candidates may not meet every single role requirement listed above. If your experience looks a little different from our requirements but you believe that you can still bring value to the role, we'd love to see your application!

We acknowledge that many candidates may not meet every single role requirement listed above. If your experience looks a little different from our requirements but you believe that you can still bring value to the role, we’d love to see your application!​

We’re Criteo, the Commerce Intelligence Platform. Criteo helps businesses turn shopper signals into commerce outcomes while delivering more relevant experiences for shoppers. We use proprietary commerce intelligence and AI decisioning to drive relevance for shoppers and performance for businesses.
At Criteo, our culture is as unique as it is diverse. From our offices across the globe or from the comfort of home, our 3,600 Criteos collaborate together to build an open, impactful, and forward-thinking environment.
We foster a workplace where everyone is valued, and employment decisions are based solely on skills, qualifications, and business needs—never on non-job-related factors or legally protected characteristics.