Roche's Research and Early Development organisations at Genentech (gRED) and Pharma (pRED) need to leverage advances in AI, data, and computational sciences to accelerate drug discovery and development. Seamless data sharing and access to models across gRED and pRED are essential to maximizing these opportunities. The new Computational Sciences Center of Excellence (CoE) aims to harness the transformative power of data and Artificial Intelligence (AI) to assist scientists in delivering more innovative and transformative medicines.
Requirements
- Proficient in Python, with hands-on experience using modern frameworks for deep learning and GenAI, such as PyTorch, Hugging Face Transformers, LangChain, or Llama-Index.
- Good understanding of machine learning algorithms, model evaluation techniques, and performance optimization, with a knowledge of deploying LLMs in data-intensive settings.
- Skilled in cloud platforms (AWS, GCP, Azure), version control systems (Git, DVC, MLflow), CI/CD pipelines, and SQL for relational database management.
- A public portfolio of projects available on GitHub/GitLab.
- Experience in deploying machine learning applications at scale, preferably in R&D or data-intensive environments.
- Continuously updated on advancements in LLMs and GenAI.
Responsibilities
- Design, develop, and deploy cloud-first, API-driven machine learning applications for data search, insights, and protocol generation and review platforms.
- Leverage large language models (LLMs) to improve contextual search, data retrieval, and scientific research efficiency through advanced prompt engineering, retrieval augmented generation, and fine-tuning techniques.
- Develop and refine LLMs tailored for protocol generation and review workflows, driving innovation in GenAI applications to streamline R&D processes.
- Collaborate with data engineers, software engineers, and architects to integrate ML models effectively within the internal data ecosystem.
- Monitor, validate, and optimize ML applications to ensure high-quality outputs, performance scalability, and a seamless user experience.
- Partner with research teams to identify needs, exchange insights, and deliver solutions that address evolving R&D requirements.
Other
- Work closely with key stakeholders to deliver impactful machine learning solutions that benefit our broader R&D community.
- Working closely with researchers, scientists, and engineers, you will bring a harmonious approach and technical rigor to projects that fulfill our scientific teams' needs.
- Be a collaborative problem-solver with a strong sense of ownership, capable of partnering with interdisciplinary teams to deliver impactful solutions.
- Onsite presence on our South San Francisco campus is expected for at least 3 days a week.
- A record of scientific excellence, as evidenced by at least one publication in a scientific journal or conference.
