Principal Scientist Computational Chemistry

About the role

We are seeking an accomplished and self-motivated principal scientist with exceptional interpersonal and problem-solving skills to join our team.

At Apheris, we power federated data networks in life sciences to address the data bottleneck in training highly performant ML models. Publicly available molecular datasets are insufficient to train high-quality ML models that meet industry requirements. Our product hosts networks where biopharma organizations collaboratively train higher quality models on their combined data. Our federated computing infrastructure, with built-in governance and privacy controls, ensures that data IP and ownership always stay with the data custodians.

We focus on two key platforms: one around protein structure and binding predictions ( ), and one around ADMET predictions ( ). We seek a Principal Scientist to lead the scientific direction and execution of our ADMET network. This high-impact role aims to transform pharmaceutical collaboration and accelerate drug discovery, ultimately improving patient lives globally.

You will lead scientific interactions with pharmaceutical partners and work closely with our leadership on scientific strategy. This role involves developing, applying, and advancing state-of-the-art deep learning models. You will serve as the scientific authority on data pipelines, harmonization, and experimental design, mentoring engineers and researchers. Deep understanding of drug discovery, ADMET, pharmacokinetics, and toxicity is essential, along with fluency in communicating complex concepts to external stakeholders.

While this role focuses on ADMET, future networks in drug discovery are planned, offering broader scientific impact. If you want to be part of a mission-driven team building cutting-edge AI systems for life sciences, this role is ideal.

About you

What you will do

1. Drive the scientific approach for developing and benchmarking state-of-the-art ADMET models on distributed, external data.
2. Influence strategic decisions on data strategy, experimental design, and benchmarking.
3. Lead scientific efforts for our ADMET network, defining data preprocessing, selection, harmonization, and benchmarking strategies, and mentor team members.
4. Contribute to publications or open-source projects where relevant.
5. Design experimental methodologies for benchmarking ADMET prediction models.
6. Develop a technical understanding of the Apheris product related to ADMET use-cases, owning experimentation methodologies, and creating roadmaps and plans.
7. Lead data preparation efforts in ADMET, demonstrating measurable improvements in model performance and impact.

You should apply if

• You have successfully employed cutting-edge machine learning models in drug discovery.
• You possess expert knowledge of models like MD, FEP, ADMET, multitask learning, and GNN approaches.
• You are highly collaborative and an effective communicator.
• You are organized with strong time management skills.
• You are proactive, driven, and promote a culture of innovation and inclusion.
• You hold a PhD in computational chemistry, cheminformatics, computational biology, or bioinformatics, with at least 5 years of relevant experience.
• You have led strategic scientific initiatives in drug discovery and have experience executing ambitious modeling plans.
• You possess expert knowledge of ADMET models and their role in drug discovery, with deep experience in data and assay protocols.
• You have strong skills in machine learning, scientific computing, data analysis, and cloud computing.

Bonus points if

• You have built or trained ADMET ML models using architectures like GNNs or Transformers.
• You have experience with federated learning, privacy-preserving ML, or secure model training.
• You have published in top-tier ML or biology journals/conferences.
• You have experience working in consortia and aligning with external stakeholders.
• You have contributed to open-source ML or cheminformatics tools.

What we offer you

• Competitive compensation, including early-stage virtual share options.
• Remote-first work environment.
• Benefits like wellbeing budgets, mental health support, co-working stipends, and learning budgets.
• Team lunches and social events.
• Generous holiday allowance.
• Quarterly meet-ups at our Berlin HQ or elsewhere in Europe.
• A diverse, mission-driven team committed to using AI for good.
• Opportunities for personal and professional growth and role shaping.

About Apheris

Apheris enables federated life sciences data networks, overcoming data silos due to IP and privacy concerns. Our product allows organizations to collaboratively train high-quality models on shared data, focusing on structural biology and ADMET.

Logistics

Our interview process includes three phases:

1. Initial Screening: A video call to explore fit, skills, and questions.
2. Deep Dive: An assessment by a domain expert on relevant skills and scenarios.
3. Final Interview: Up to three hours with our founders to discuss culture and meet future colleagues.

#J-18808-Ljbffr