
Genentech, a member of the Roche Group, is offering a specialized 12-week internship within its Translational AI & Learning (TRAIL) group. This group is part of the AI for Biology & Translation (AIBT) department, which focuses on utilizing cutting-edge machine learning to decode complex biological systems and accelerate drug discovery.
This role is ideal for PhD students who want to bridge the gap between advanced ML engineering and real-world biomedical impact in South San Francisco.
The TRAIL Internship Experience
Interns will work at the intersection of high-dimensional data and predictive modeling. Key project areas include:
- Foundation Models: Applying and adapting large-scale biological foundation models to complex disease biology.
- Multi-Modal Integration: Developing predictive models for disease phenotypes using diverse datasets (genomics, proteomics, etc.).
- Interpretable ML: Building models that don’t just predict, but help scientists understand “why,” supporting new drug target discovery.
- Benchmarking: Designing rigorous frameworks to test how well biological ML models generalize to real-world patient data.
Program Details
- Duration: 12 weeks, Full-time (40 hours/week).
- Timeline: Starting May or June 2026.
- Compensation: Paid internship with a location-based stipend.
- Environment: Onsite in South San Francisco, working with top-tier computational biologists and ML engineers.
Candidate Requirements
| Category | Requirement |
| Education | Current PhD student (enrolled). |
| Majors | CS, Computational Biology, Bioinformatics, Statistics, Data Science, EE, or Math. |
| Technical Skills | Strong Python programming; prior research in ML/AI or Genomics. |
| Preferred Tools | PyTorch (preferred) or TensorFlow; experience with single-cell/sequence modeling. |
Core Responsibilities
- Model Development: Implement ML models using large-scale genomics datasets.
- Data Curation: Preprocess and analyze high-dimensional data (e.g., single-cell and multi-modal datasets).
- Rigorous Evaluation: Design experiments including baselines and ablations to ensure model reliability.
- Communication: Translate complex scientific questions into modeling approaches and present results to the broader team.
How to Apply
- Job ID: 202601-101276
- Location: South San Francisco, CA (Onsite)
- Key Focus: Highlight your experience with deep learning frameworks and your ability to handle large-scale, reproducible workflows.
