Principal ML Scientist – Predictive Toxicology

This is a fully remote position, open to applicants in Germany.

📋 Description

• Take charge of the company's growth into predictive toxicology and quantitative biology.

• Establish the scientific strategy for in silico toxicology and quantitative biology processes across distributed networks.

• Identify significant endpoints, assays, and modeling techniques for genuine drug-discovery decisions.

• Assess how relevant data and methodologies can yield both scientific and commercial benefits.

• Utilize federated learning with partner data to create high-performing models.

• Incorporate scientific workflows into the platform, enabling customers to execute them at scale.

• Collaborate with industrial partners to integrate models within drug-discovery pipelines.

• Lead scientific discussions with customers and partners, overseeing scope, evaluation, delivery, and adoption.

• Influence the product roadmap based on scientific and commercial requirements.

• Stay actively engaged in modeling while providing scientific guidance and mentoring others.

⛳️ Requirements

• Strong foundational knowledge in deep learning for molecular AI.

• Experience with graph neural networks, message-passing, and transformer-based models for molecular property modeling.

• Familiarity with toxicity assessment issues in drug discovery, including endpoints like DILI, cytotoxicity, or micronucleus/genotoxicity imaging readouts.

• Proven experience in developing predictive models and promoting the use of toxicity models in actual drug-discovery programs or industrial R&D pipelines.

• Working knowledge of RNA-seq, toxicity screens, and image-based screens in pharmaceutical high-throughput screening (HTS) and compound triage.

• Capability to establish a scientific vision, manage a scientific agenda, and lead technical and customer discussions independently.

• Ability to remain engaged in modeling while directing scientific efforts and mentoring colleagues.

• PhD or equivalent experience in computational biology, cheminformatics, toxicology, machine learning, or a related field.

• Over 6 years of experience applying machine learning to drug discovery or life science challenges.

• Nice to have: experience with federated learning, privacy-preserving machine learning, or multi-party training environments.

• Nice to have: evidence of prospective validation and application of predictive toxicity models in active drug-discovery programs.

• Nice to have: experience delivering production-grade models in regulated environments such as enterprises, pharmaceuticals, or biotech, along with a relevant publication record.

• Nice to have: experience in multi-omics and high-content imaging, such as cell painting.

• Nice to have: familiarity with Tox21, ToxCast, LINCS/L1000, and adverse outcome pathways.

🏝️ Benefits

• Competitive compensation package, including early-stage virtual share options.

• Remote-first work culture – operate from your ideal location.

• Wellbeing budget to support your health.

• Mental health assistance available.

• Work-from-home budget to facilitate your productivity.

• Co-working stipend to enhance your working environment.

• Learning budget for professional development.

• Generous holiday entitlement.

• Office Days at our Berlin headquarters or another European location (three times a year).

• Join a high-caliber, execution-focused team with experience from leading organizations.