Syndata is seeking a Data Scientist / Machine Learning Engineer to join our research and engineering team working on LeakPro 2.0 – our next-generation framework for evaluating privacy leakage, synthetic data utility, and AI robustness.

The role can be part-time (minimum 50%) or full-time, and applicants are encouraged to specify their preferred workload.

This is an opportunity to contribute to cutting-edge work at the intersection of synthetic data, generative AI, privacy-preserving technologies, and machine learning engineering.

Key Responsibilities

• Design and implement empirical evaluation pipelines for privacy leakage, robustness and utility testing.

• Develop, train and benchmark generative models (GANs, VAEs, diffusion models, LLM-based generators) across multiple data modalities (tabular, image, text, time-series).

• Investigate and execute privacy attack vectors such as membership inference, singling out, attribute inference and linkability.

• Contribute to defining and improving the scientific foundations behind LeakPro 2.0.

• Build modular, reproducible ML systems and research code using modern engineering practices.

• Collaborate with legal and policy teams to translate technical results into insights relevant for GDPR, the AI Act and the EHDS.

Required Technical Qualifications

• Strong statistical grounding and experience with empirical research or experiment design.

• Practical experience generating synthetic data in one or more modalities (tabular, image, text, time-series).

• Proficiency with modern generative models: GANs, VAEs, diffusion models, or LLM-based synthesizers.

• Knowledge of privacy-preserving methods, including:

  • Differential Privacy (e.g., DP-SGD)

  • Federated Learning

  • Secure Multi-Party Computation

  • Zero-Knowledge Proofs (nice to have)

• Understanding of privacy risks and attack techniques.

• Experience training and evaluating machine learning models.

Software & Engineering Skills

• Advanced Python programming skills.

• Experience with PyTorch or TensorFlow.

• Ability to write clean, modular, well-documented research code.

• Familiarity with benchmarking frameworks, automated testing and reproducible ML pipelines.

• Strong version control discipline (Git).

  
  

Collaboration & Communication

• Ability to explain complex technical concepts to non-expert audiences.

• Experience working in cross-disciplinary research settings.

• Comfortable interfacing with legal and compliance topics related to AI.

Who We’re Looking For

A candidate with a strong analytical mindset, interest in rigorous experimentation, and the ability to move between theoretical reasoning and practical engineering. You enjoy building high-quality research tools, thinking deeply about privacy and risk, and contributing to a product with real societal impact.

Employment Details

• Workload: 50–100% (state your preference in the application)

• Location: Remote-first and/or Stockholm optional - but living in Sweden is a must.

• Start date: As soon as possible

• Contract: Project-based employment

• Compensation: Competitive, based on experience

We have no interest in consultants or recruitment firms for this opening.