Postdoctoral Fellow - AI for Drug Discovery

2w2 weeks ago

Genentech

South San Francisco, US · Full-time · $110,000 – $120,000

About this role

Genentech seeks a postdoctoral fellow for the AI for Drug Discovery department in the Computational Sciences Center of Excellence. This Frontier Research role focuses on scalable methods for sequential experimental design in machine learning systems. Applications target hyperparameter optimization, model selection, and compute-aware adaptation.

Combine theoretical ideas from Bayesian optimization, meta learning, and multi-fidelity learning to create decision procedures. Exploit partial or intermediate observations to improve with experience across tasks. Aim to make experimentation with large models more principled, sample-efficient, and reliable.

Collaborate with colleagues on foundation models and agents for drug discovery. Translate theory into practical algorithms for training, adapting, and deploying large models in scientific discovery. Build and release research artifacts like open-source code, benchmark suites, and model weights.

Join Genentech’s Postdoctoral Program to conduct world-class research and publish in top-tier journals. Receive mentorship to develop into an independent scientist and leader. Elevate your career in biotechnology with a prestigious community of early career scientists.

Requirements

Ph.D. in computer science, statistics, physics, or related computational field
Strong publication record and experience contributing to research communities, including journals and conferences in machine learning or statistics
Demonstrated expertise in one or more relevant areas, such as Bayesian optimization, active learning, reinforcement learning, bandits, adaptive experimentation, control, and online optimization
Independent, motivated, and highly collaborative, with interests in both the theory and practice of adaptive experimentation
Experience with pretraining, post-training, or inference-time adaptation of foundation models such as large language models

Responsibilities

Develop new theory and algorithms at the intersection of sequential decision making, experimental design, and large-scale machine learning
Translate theoretical ideas into practical algorithms motivated by challenges in training, adapting, and deploying large models for scientific discovery
Build and release research artifacts, including open-source code, benchmark suites, and model weights
Collaborate with colleagues working on foundation models and agents for drug discovery to ground methods in key scientific applications
Publish in leading machine learning conferences or journals