Senior ML Engineer – ADMET, Toxicity Networks

Apheris

. Own model pipelines end to end, from partner data landing through data preparation, training, evaluation and release of benchmarked model weights for ADMET and toxicity endpoints .

Posted 9/29/2026full-timeBerlin • GermanySeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in building and managing machine learning systems, particularly in molecular ML and cheminformatics, with a strong focus on data validation, reproducibility, and production-grade model delivery in regulated environments. Proficient in translating scientific requirements into actionable pipeline behaviors while ensuring data integrity and performance reporting.

Highest-signal resume keywords

Machine Learning Systems DevelopmentMolecular ML or CheminformaticsPython ProgrammingML Ops or ML InfrastructureProduction-Grade Model Delivery

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Machine LearningPythonData ValidationModel EvaluationKubernetesRDKitPerformance ReportingData Leakage PreventionVersion ControlModular Code Development

Soft Skills

CollaborationProblem-SolvingCommunication

Tools & Technologies

PyTorchKubernetesChEMBLTox21ToxCast

Industry Keywords

ADMETToxicity PredictionBiotechPharmaceuticalFederated Learning

Tech Stack

Tools & technologies

KubernetesPythonPyTorch

About the role

Key responsibilities & impact

Own model pipelines end to end, from partner data landing through data preparation, training, evaluation and release of benchmarked model weights for ADMET and toxicity endpoints
Build schema and data-contract validation, actionable-error validators, and QC/profiling reports usable without access to partners' raw data
Make federated runs reproducible and auditable through versioned configurations, pinned data snapshots and model provenance
Build leakage-safe splitting, held-out benchmarks and honest performance reporting
Turn prototypes and research code into tested, modular systems and hand them to engineering for scaling into Foundry
Translate scientific requirements into pipeline behavior with the science team
Surface data or modeling risks early to partners and internal teams

Requirements

What you’ll need

5+ years building ML systems in Python, with version control, tested modular code, code review, and interfaces other people can use
Hands-on molecular ML or cheminformatics using RDKit, fingerprints and descriptors, or graph/transformer models, applied to property, activity or toxicity prediction
Experience building training and evaluation pipelines that other people run, not one-off notebooks
Understanding of molecular ML failure modes including data leakage, split design, applicability domain, dataset shift, and over-optimistic benchmarks
Ability to write validators and data contracts under partial visibility, where raw data cannot be inspected
Comfortable with PyTorch or an equivalent modern ML stack
Ability to translate ambiguous scientific requirements into defined, testable pipeline behavior
Federated learning, privacy-preserving ML, or other multi-party training environments
ML Ops or ML infrastructure experience, particularly Kubernetes-based training, evaluation or deployment workflows
Production-grade model delivery in regulated, enterprise, pharmaceutical or biotech settings
Familiarity with public ADMET, toxicity and bioactivity data resources including ChEMBL, Tox21 and ToxCast
Open-source contributions or a publication record in cheminformatics, molecular ML, or applied machine learning

Benefits

Comp & perks

Industry-competitive compensation, including early-stage virtual share options
Remote-first working - work where you work best
Wellbeing budget
Mental health support
Work-from-home budget
Co-working stipend
Learning budget
Generous holiday allowance
Office days at our Berlin HQ or a different European location (3x per year)