Data Scientist - Extensions

Overview

This opportunity is for a Data Scientist focused on Extensions, based in Germany and handled by a partner company that manages the application process and subsequent hiring steps. The role is centered on improving advanced AI systems that support enterprise decision-making. You will work on challenging structured-data problems and help raise the performance of large tabular models across different industries and practical business scenarios.

The position combines research and production work. You will turn exploratory ideas into dependable, production-ready solutions that create measurable value for enterprise customers. The environment is technically demanding, fast-paced, and strongly research-oriented, with the chance to contribute to foundational AI capabilities used by large organizations to make quicker and better decisions.

Key Responsibilities

• Develop and refine modern data science methods that lift predictive quality across large enterprise datasets and a variety of prediction problems.
• Build, maintain, and improve production-grade Python modules with attention to reliability, scalability, and reuse.
• Study the behavior of real-world enterprise data and create approaches that keep models strong under missing values, imbalance, and data drift.
• Plan and execute robust experiments, create benchmarks, and measure improvements using sound statistical evaluation.
• Work on structured machine learning tasks such as classification, regression, ranking, and forecasting.
• Partner with research and engineering colleagues to interpret model behavior and convert insights into product-level improvements.
• Collaborate with Applied AI Engineers to test solutions on customer datasets and turn validated ideas into deployable features.
• Contribute to documentation, internal tools, and engineering practices that improve reproducibility and team knowledge sharing.

Requirements

• At least 5 years of experience in data science, machine learning, or applied ML engineering.
• Strong command of Python, along with practical use of pandas, NumPy, and scikit-learn.
• Deep working knowledge of tree-based and gradient-boosting approaches such as XGBoost, LightGBM, CatBoost, and similar frameworks.
• Solid understanding of tabular-data challenges like missingness, class imbalance, high-cardinality variables, and distribution shift.
• Proven ability to design experiments, build benchmarks, and draw reliable conclusions from imperfect or noisy data.
• Capability to independently take a project from research concept through production implementation.
• Strong analytical reasoning and a problem-solving mindset focused on measurable outcomes and empirical validation.
• Experience with structured prediction problems in areas like finance, healthcare, supply chain, retail, or industrial settings is an advantage.
• Exposure to tabular foundation models such as TabPFN or CARTE is a strong plus.
• Familiarity with DuckDB, Polars, or similar in-process analytics tools is beneficial.
• Experience in competitive data science settings such as Kaggle or DrivenData, or contributions to ML libraries, is helpful.
• Ability to read machine learning research and apply it effectively in practical implementations.

Benefits

• Attractive compensation package that includes salary and equity.
• Full health coverage for the employee and dependents.
• Paid parental leave for all parents, including adoptive and surrogate families.
• Relocation assistance for those joining office-based locations.
• A mission-led, low-ego environment built around ownership, teamwork, and impact.
• The chance to work on foundational AI systems influencing enterprise decision-making at scale.
• High autonomy in a technically challenging, research-focused setting.

Application and Hiring Process

This role is processed through an AI-supported matching workflow. Applications are screened against the role’s core requirements to produce a shortlist, which is then shared with the hiring company. Final hiring decisions, including interviews and assessments, are handled directly by the employer’s internal team.

Privacy and Data Handling

By applying, you consent to the processing of your personal data for candidate evaluation and the sharing of relevant information with the hiring employer. This is based on legitimate interest and pre-contractual measures under applicable data protection law, including GDPR. You may request access to, correction of, deletion of, or objection to the processing of your data at any time.

The recruitment process may use AI-assisted tools to support tasks such as application review, resume analysis, response evaluation, and identification of potential inconsistencies or verification signals. These tools assist the recruiting team but do not replace human judgment. Final hiring decisions are made by people.