Freenome is developing tests to detect cancer using a standard blood draw by combining a multiomics platform with machine learning. The company needs to develop and deploy infrastructure to support the development of deep learning models using massive-scale genomic data, which presents significant challenges for current training paradigms.
Requirements
- 5+ years of post-MS industry experience working on developing AI/ML software engineering pipelines.
- Proficiency in a general-purpose programming language: Python (preferred), Java, Julia, C, C++, etc.
- Strong knowledge of ML and DL fundamentals and hands-on experience with machine learning frameworks such as PyTorch, TensorFlow, Jax or Scikit-learn.
- In-depth knowledge of scalable and distributed computing platforms that support complex model training (such as Ray or DeepSpeed) and their integration with ML developer tools like TensorBoard, Wandb, or MLflow.
- Experience with cloud platforms (e.g., AWS, Google Cloud, Azure) and how to deploy and manage AI/ML models and pipelines in a cloud environment.
- Understanding of containerization technologies (e.g., Docker) and computing resource orchestration tools (e.g., Kubernetes) for deploying scalable ML/AI solutions.
- Proven track record of developing and optimizing workflows for training DL models, large language models (LLMs), or similar for problems with high data complexity and volume.
Responsibilities
- Implement and refine DL pipelines on distributed computing platforms enhancing the speed and efficiency of DL operations including model training, data handling, model management, and inference.
- Collaborate closely with ML scientists and software engineers to understand current challenges and requirements and ensure that the DL model development pipelines you create are perfectly aligned with scientific goals and operational needs.
- Continuously monitor, evaluate, and optimize DL model training pipelines for performance and scalability.
- Stay up to date with the latest advancements in AI, ML, and related technologies, and quickly learn and adapt new tools and frameworks, if necessary.
- Develop and maintain robust and reproducible DL pipelines that guarantee that DL pipelines can be reliably executed, maintaining consistency and accuracy of results.
- Drive performance improvements across our stack through profiling, optimization, and benchmarking.
- Implement efficient caching solutions and debug distributed systems to accelerate both training and evaluation pipelines.
Other
- MS or equivalent experience in a relevant, quantitative field such as Computer Science, Statistics, Mathematics, Software Engineering, with an emphasis on AI/ML theory and/or practical development.
- Act as a bridge facilitating communication between the engineering and scientific teams, documenting and sharing best practices to foster a culture of learning and continuous improvement.
- Excellent ability to work effectively with cross-functional teams and communicate across disciplines.
- This can be a hybrid role based in our Brisbane, California headquarters (2-3 days per week in office), or remote.
