TL;DR
Postdoctoral Fellow (AI): Developing scalable sequential experimental design methods for machine learning with an accent on Bayesian optimization, meta-learning, and multi-fidelity learning. Focus on building algorithmic frameworks to improve sample efficiency and reliability in training foundation models for drug discovery.
Location: Must be based in or able to relocate to South San Francisco, California
Salary: $110,000–$120,000
Company
A pioneering biotechnology company and member of the Roche group, dedicated to developing transformative medicines for serious diseases.
What you will do
- Develop new theory and algorithms at the intersection of sequential decision-making and large-scale machine learning.
- Translate theoretical research into practical algorithms for training and adapting foundation models.
- Collaborate with interdisciplinary teams to apply methods to drug discovery challenges.
- Publish findings in leading machine learning conferences and journals.
- Release open-source research artifacts, including code, benchmark suites, and model weights.
Requirements
- PhD in computer science, statistics, physics, or a related computational field.
- Strong publication record in machine learning or statistics.
- Expertise in at least one area: Bayesian optimization, active learning, reinforcement learning, or adaptive experimentation.
- Ability to work independently and collaboratively in a research-heavy environment.
Nice to have
- Experience with pretraining, post-training, or inference-time adaptation of foundation models (e.g., Large Language Models).
Culture & Benefits
- Competitive salary with fully funded research expenses.
- Access to world-class seminars, professional development workshops, and networking opportunities.
- Opportunity to build a robust scientific network in the biotechnology sector.
- Comprehensive employee benefits including health insurance and paid time off.
- Supportive environment dedicated to becoming an independent scientific leader.