Applied Tabular AI Intern (AI)

Location: Onsite at hirify.global HQ in Paris, France

Company

hirify.global is a tabular AI company founded by the creators and stewards of scikit-learn, developing Skore and maintaining an ecosystem of open-source data science tools.

What you will do

• Prepare and co-run client demos, workshops, hackathons, and proof-of-concept sessions with senior engineers.
• Build proofs of concept on real client data through scoping, feature engineering with skrub, baselines with scikit-learn, evaluation, leakage detection with Skore, and actionable client readouts.
• Use the Skore agent and LLM coding agents in client engagements, identifying failure modes and creating reusable skills and checks.
• Reproduce field issues and contribute them to Skore, skrub, or scikit-learn.

Requirements

• Final-year or gap-year student at a leading engineering school or research master's program.
• Strong scientific Python skills and completed end-to-end tabular projects beyond coursework, such as competitions, research, or side projects.
• Strong understanding of evaluation, including correct cross-validation, leakage, calibration, and business metrics.
• Ability to explain models to non-technical stakeholders and work comfortably in front of clients, in French and English.
• Curiosity about tabular foundation models and LLM agents for data science.
• Eligible for a tripartite internship agreement and authorized to work full-time in France for the entire six-month placement.

Nice to have

• Daily use of a coding agent, open-source contributions, or a self-completed deployment.

Culture & Benefits

• Six-month internship in the Services team, focused on applied AI and field engineering.
• Work on live engagements and real client data with a senior engineer alongside you.
• Exposure to enterprise AI adoption, agentic data science, and open-source contributions reviewed by maintainers.
• Monthly internship stipend; amount to be determined.
• Flexible start date; a successful internship may lead to a full-time Forward Deployed Engineer, Applied AI, or Solutions Engineer role.

Hiring process

• Submit a CV, a readable link such as GitHub, a project, or a competition result, and five to ten lines about the most interesting tabular data problem you worked on and what you learned; no cover letter.
• The approximately two-week process includes a 30-minute call with the VP Services, a one-hour technical case on a tabular dataset, and a conversation with an FDE or Solution Architect.