Grailbio1mo ago

Machine Learning Infrastructure Engineer - #4694

Edison · Menlo ParkFull-Timemid

Data ScienceOtherDevOps & InfrastructureMachine LearningMachine Learning Infrastructure EngineerData & AI

0 views0 saves0 applied

Apply Now

Quick Summary

Overview

Our mission is to detect cancer early, when it can be cured. We are working to change the trajectory of cancer mortality and bring stakeholders together to adopt innovative, safe,

Technical Tools

Data ScienceOtherDevOps & InfrastructureMachine LearningMachine Learning Infrastructure EngineerData & AI

Our mission is to detect cancer early, when it can be cured. We are working to change the trajectory of cancer mortality and bring stakeholders together to adopt innovative, safe, and effective technologies that can transform cancer care.

We are a healthcare company, pioneering new technologies to advance early cancer detection. We have built a multi-disciplinary organization of scientists, engineers, and physicians and we are using the power of next-generation sequencing (NGS), population-scale clinical studies, and state-of-the-art computer science and data science to overcome one of medicine’s greatest challenges.

GRAIL is headquartered in the bay area of California, with locations in Washington, D.C., North Carolina, and the United Kingdom. It is supported by leading global investors and pharmaceutical, technology, and healthcare companies.

For more information, please visit grail.com

GRAIL is seeking a Staff Machine Learning Infrastructure Engineer for the Research Platform Engineering team. This is a software engineering role, charged with building and supporting systems executing machine learning and other analysis workflows on controlled data. You will empower computational biologists, data scientists, and statisticians in their quest to develop and refine powerful diagnostic products, by enabling efficient and flexible exploratory research and classifier development, and smoothing the productionization of their work.

The ideal candidate will bring a passion for reliable software infrastructure, distributed computing, reproducible research, and general problem-solving. Due to the highly connected nature of this position, the candidate should be a strong communicator with experience working with multidisciplinary teams.

This is a hybrid role based in Menlo Park, CA (moving to Sunnyvale, CA in Fall 2026). Our current hybrid policy requires on-site presence at least 40% of the time, including key in-person collaboration days. At our Menlo Park campus, Tuesdays and Thursdays are the key days where we encourage on-site presence to engage in events and on-site activities.

Partner with research teams to identify computational pain points or limitations in performing computational experiments and analyses.

Design, build, and evolve software which usefully extends research capabilities, including infrastructure for distributed ML training and evaluation on large controlled genomic datasets.

Develop tools and processes that ensure GxP-compliant testing, patchability, and inference reproducibility for classifiers that are promoted to production use.

Develop and maintain the research team’s software environment, including tools to assess the health, performance, and cost of the system.

These summarize the role’s primary responsibilities and are not an exhaustive list. They may change at the company’s discretion.

5+ years of experience developing software supporting machine learning, scientific computing, or large-scale data processing systems

Strong programming skills in Python and a systems-level language such as Golang (preferred), Java, C#, C++, etc.

Experience working with modern machine learning frameworks such as PyTorch or TensorFlow

Experience with Distributed Computing paradigms (Spark, Ray, Flink, Beam, etc.)

A commitment to high-quality professionally engineered software

Strong communication skills with the ability to help developers from a wide range of software development backgrounds

BS in Computer Science, Engineering, Bioinformatics, or a related field, or equivalent practical experience

Good understanding of container orchestration through Docker and cloud technologies.

Experience with scientific computing tools: NumPy, Jupyter, R Notebook, etc.

Experience with techniques used in modern AI (including LLM) training

Experience with whole genome sequencing, whole exome sequencing, bisulfite sequencing, and/or whole transcriptome sequencing data

Practical experience setting up continuous integration systems, along with expertise in at least one build tool (e.g. Bazel (preferred), Buck, Maven, Gradle)

Familiarity with AWS services, best practices, and security

Advanced degree (MS or PhD) in computer science, engineering, bioinformatics or a related discipline