Senior Machine Learning Scientist

Summary

Join Freenome, a high-growth biotech company, as a Senior Machine Learning Scientist. You will develop algorithms and pipelines to analyze electronic health records (EHR) and other real-world data (RWD) for cancer detection. Leverage your expertise in AI, natural language processing, and multimodal data analysis to extract insights from complex datasets. Collaborate with a multidisciplinary team to design and drive research experiments. This hybrid role can be based in Brisbane, California, or remote. Contribute to Freenome's mission of reducing cancer mortality through accessible early detection. The ideal candidate possesses extensive experience with deep learning (DL) and large language models (LLMs), and a strong publication record.

Requirements

• PhD or equivalent research experience with an AI/DL emphasis and in a relevant, quantitative field such as Computer Science, Statistics, Mathematics, Computational Biology with a strong track record in natural language processing
• 3+ years of postdoc or post-PhD industry experience achieving impactful results using relevant modeling techniques
• Expertise, demonstrated by research publications or industry achievements, in applied machine learning, deep learning and complex multimodal data modeling
• Extensive experience in working with DL models, LLMs, and multimodal foundation models
• Extensive experience with training paradigms like supervised learning, self-supervised learning, and contrastive learning
• Practical and theoretical understanding of fundamental ML models like generalized linear models, kernel machines, decision trees and forests, neural networks
• Solid grasp of NLP techniques, including but not limited to named entity recognition (NER), text summarization, and question answering
• Proficiency in a general-purpose programming language: Python (preferred), Java, C, C++, etc
• Proficiency in one or more ML frameworks such as Pytorch, Tensorflow, and Jax; LLM specific frameworks like LangChain; and ML platforms like Hugging Face
• Experience in ML analysis and developer tools like TensorBoard, MLflow or Weights & Biases
• Excellent ability to communicate across disciplines, work collaboratively, and make progress in smaller steps via experimental iterations
• A passion for innovation and demonstrated initiative in tackling new areas of research

Responsibilities

• Independently pursue cutting edge research using advanced AI algorithms and LLMs to analyze EHR data, extracting relevant information for cancer detection
• Stay abreast of the latest developments in LLMs for natural language processing (NLP) and apply these models directly or after fine-tuning for clinical data analysis at Freenome
• Design and develop AI pipelines, working closely with ML engineers, to increase efficiency of biomedical and clinical data extraction, processing, and interpretation
• Collaborate with clinical data scientists and informaticians to understand data requirements and ensure the accuracy and relevance of AI-generated insights
• Collaborate with machine learning scientists to integrate EHR data with non-EHR d data sources, such as genomics or proteomics data, building robust multimodal models for cancer detection
• Lead the development and optimization of algorithms and models that leverage diverse data types, ensuring high accuracy and reliability and predictions
• Take a mindful, transparent, and humane approach to your work

Preferred Qualifications

• Experience in leveraging LLMs for EHR or other RWD data in healthcare or diagnostics
• Demonstrated experience in integrating diverse data types with EHR data for multimodal analysis
• Experience with containerized cloud computing environments such as Docker in GCP, Azure, or AWS
• Experience in a production software engineering environment, including the use of automated regression testing, version control, and deployment system

Benefits

• Pre-IPO equity
• Cash bonuses
• A full range of medical, financial, and other benefits
• This role can be a hybrid role based in our Brisbane, California headquarters (2-3 days per week in office), or remote