Senior​/Principal Machine Learning Scientist, Scientific Reasoning Models, AI Drug Discovery

The Position

A healthier future. It’s what drives us to innovate. To continuously advance science and ensure everyone has access to the healthcare they need today and for generations to come. Creating a world where we all have more time with the people we love. That’s what makes us Roche.

Advances in AI, data, and computational sciences are transforming drug discovery and development. Roche’s Research and Early Development organisations at Genentech (gRED) and Pharma (pRED) have demonstrated how these technologies accelerate R&D, leveraging data and novel computational models to drive impact. Seamless data sharing and access to models across gRED and pRED are essential to maximising these opportunities. The new Computational Sciences Center of Excellence (CoE) is a strategic, unified group whose goal is to harness the transformative power of data and Artificial Intelligence (AI) to assist our scientists in both pRED and gRED to deliver more innovative and transformative medicines for patients worldwide.

At Roche's AI for Drug Discovery (AIDD) group (Prescient Design), we are revolutionizing drug discovery with cutting‑edge machine learning (ML) techniques. We are seeking a Senior or Principal Machine Learning Scientist to join the Foundation Models team within Prescient Design (gRED). In this role, you will drive the development of our internal reasoning Large Language Models (LLMs) and enable it to succeed at drug discovery tasks, including biomolecular design.

You will work at the intersection of engineering and research, designing and scaling large machine learning systems.

In this role, you will:

• Technical Leadership & Strategy:
  Lead the design and evolution of scientific reasoning systems, setting technical direction for model architectures, training strategies, and evaluation methodologies.
• Model Capability & Improvement:
  Define and execute approaches to systematically improve model performance on scientific tasks, including long‑horizon reasoning and complex decision‑making.
• Domain Translation:
  Translate biological and chemical domain knowledge into machine learning objectives, training signals, and evaluation criteria, working closely with domain experts.
• Scalable Systems & Engineering:
  Architect and improve large‑scale distributed machine learning systems, ensuring robustness, efficiency, and reproducibility across training and evaluation workflows.
• Research‑to‑Production Impact:
  Partner with researchers and cross‑functional teams to move models from research prototypes to production‑ready systems that support active discovery programs.

• You are the primary driver of technical implementation for scientific reasoning, translating high‑level research goals into robust training code.
• You own the end‑to‑end integrity of large‑scale training runs, from data orchestration to the development of rigorous reasoning benchmarks.
• You act as a technical mentor to junior staff and interns, fostering a culture of engineering excellence and rapid experimentation.

• You help define the long‑term technical roadmap for scientific reasoning models, identifying new opportunities and setting priorities across initiatives.
• You architect new initiatives that integrate diverse data modalities, guiding the technical direction of cross‑functional projects across gRED.
• You serve as a key technical authority for leadership, influencing how Genentech leverages generative AI to solve high‑stakes problems in the therapeutic pipeline.

• PhD in Computer Science, Statistics, Mathematics, Physics, or a related quantitative field.
• Experience:
  • For Senior (SE6): 0 – 2+ years of industry or post‑doc experience with a focus on deep learning.
  • For Principal (SE7): 5+ years of industry or post‑PhD experience with a demonstrated track record of technical leadership and project ownership.
• LLM Expertise:
  Extensive experience developing and training large‑scale machine learning models, including approaches to improve domain understanding, reasoning capabilities, and model alignment.
• Publication Record: A strong history of research excellence at top‑tier venues (e.g.,…