Director, Molecular AI & Federated Learning

Scorpion Therapeutics; San Francisco (CA); On-site; USD 177,000 - 282,000; Full time

Benefits offered by this job

Company bonus

401(k)

Medical plan

Dental plan

Vision plan

Life insurance

Paid time off

Well-being benefits

Job summary

Scorpion Therapeutics seeks a Director, Molecular AI & Federated Learning (TuneLab) to unite privacy-preserving federated learning with generative small-molecule design. You will lead predictive and generative models accelerating lead optimization and candidate selection.

You'll provide technical vision, mentorship, and rigorous methods for scalable federated learning and molecular AI, shaping a platform that integrates advanced ML with chemistry domain knowledge.

Qualifications

• PhD in Computer Science, Computational Chemistry, Cheminformatics, ML, computational biology, or related field.
• 5+ years post-PhD ML drug-discovery experience (prefer 8+).

Responsibilities

• Define technical direction and research agenda for federated learning and molecular AI.
• Provide principal technical leadership and mentor data scientists and engineers; guide experimental design and code reviews.
• Architect federated foundation models (Transformer and GNN-based) for large-scale pre-training.
• Develop federated optimization strategies (FedAvg/FedProx/SCAFFOLD) for non-IID data.
• Build scalable systems and simulation environments to benchmark federated strategies.

Skills

Federated learning

GNNs

PyTorch

RDKit

DeepChem

Education

PhD in relevant field

Tools

RDKit

DeepChem

PyTorch

Job description

Job Summary

• Director, Molecular AI & Federated Learning (TuneLab)
• Set the technical vision uniting privacy-preserving federated learning with generative small-molecule design; lead predictive and generative models that accelerate small-molecule lead optimization and candidate selection.
• Lead through vision, methodological rigor, and mentorship (no formal people management).

Key Responsibilities

• Define technical direction and research agenda for federated learning and molecular AI aligned to platform/portfolio priorities.
• Provide principal technical leadership and mentor data scientists and engineers; guide experimental design and review methods/code.
• Architect federated foundation models (e.g., Transformer and graph neural network–based) for large-scale federated pre-training.
• Advance semi-supervised/self-supervised learning for federated constraints (communication bottlenecks, data heterogeneity).
• Develop robust federated optimization/aggregation strategies (FedAvg, FedProx, SCAFFOLD) for non-IID data.
• Optimize scalability (memory, latency, communication cost) and build simulation environments to benchmark federated strategies.
• Architect federated multi-task learning models for shared representations across endpoints.
• Design algorithms for task/feature heterogeneity (personalization, meta-learning, gradient aggregation, regularization to prevent negative transfer).
• Create protocols for downstream adaptation/validation with per-task metrics and fairness assessment.
• Build multi-task small-molecule property prediction (ADMET, solubility, permeability, stability, off‑target liabilities).
• Design/deploy generative chemistry models (VAEs, diffusion, flow matching, autoregressive) for de novo design/optimization/scaffold hopping.
• Develop ADMET-driven multi-objective prediction–generation pipelines (Pareto-front exploration).
• Ensure synthetic feasibility via reaction‑aware generation, retrosynthetic planning integration, and collaboration with synthetic chemists.
• Learn structure–activity and representations from sparse/noisy data; apply XAI for scientific insight.
• Establish benchmarks (ChEMBL, ZINC, PubChem, proprietary data), publish/present, and uphold reproducible code/version control.

Basic Qualifications

• PhD in Computer Science, Computational Chemistry, Cheminformatics, Machine Learning, Computational Biology, or related field.
• 5+ years post-PhD ML drug-discovery experience (preference for 8+); or equivalent technical leadership/impact.

Additional Preferences

• Technical leadership without formal people-management requirement.
• Track record in generative molecular design; multi‑task/representation learning.
• Deep medicinal chemistry and ADMET optimization knowledge.
• Hands‑on federated learning, distributed optimization, privacy-preserving ML.
• Publications in top venues; expertise in GNNs/geometric deep learning.
• Organic chemistry and synthetic feasibility; fragment‑/structure‑based drug design.
• PK/PD knowledge; RDKit/DeepChem and PyTorch.
• Active learning and design–make–test–analyze; uncertainty quantification and XAI.
• Strong communication, learning agility, independent drive.

Other Information / Location

• Indianapolis, San Francisco, or Boston; up to 10% travel.

Benefits (explicitly stated)

• Company bonus (company/individual performance dependent).
• 401(k); pension; vacation; medical/dental/vision/prescription; flexible benefits; life insurance; time off/leave; well‑being benefits.

Pay Transparency (explicitly stated)

• Anticipated wage: $177,000–$281,600.