Machine Learning Reseach Engineer

What is Freenome?
Freenome is a health technology company bringing accurate, accessible and non-invasive disease screenings to you and your doctor to proactively treat cancer and other diseases at their most manageable stages. Freenome is a place where you can do your best, most meaningful work, and contribute to a whole new way of thinking about curing cancer. It’s a data-driven, diverse company at the intersection of biology, technology and medicine.

Our Mission: Equip you and your family with knowledge and tools to maintain a healthier life and prevent disease.

About the Role
At Freenome you will build models, systems, and science to detect cancer at its earliest and most actionable stages. Using machine learning tools, you will work with a multidisciplinary team of researchers and engineers to build, compare, and scale statistical solutions for early-onset cancer detection.
With the tools of machine learning, distributed systems, genetic sequencing technology, and cutting edge molecular biology your day-to-day brings early cancer detection to the people that matter: patients. You will work with an interdisciplinary team of machine learning scientists, software engineers, lab and computational biologists to do industry defining science with elegant technological solutions.

  • Engineer features from complex biological measurements
  • Implement large-scale ML experiments over biological features.
  • Including model refinement, hyperparameter-search, performance monitoring, model comparison, model serialization, and experiment reproducibility, async data loading and preprocessing. 
  • Optimize large-scale neural-network architectures to run quickly over multiple GPUs.
  • Build and immediately apply the technologies you build to improve the lives of patients,  the work of doctors, and the efficiency of healthcare
  • Development of real-time monitoring infrastructure for lengthy experiments running in a parallel execution environment

What We're Looking For
  • 3+ years of experience and proficiency with a machine learning research loop: hyper-parameter iteration, data augmentation, performance tracking, etc.
  • Demonstrated via publication record or publicly available code
  • Ability to reproduce and sweep over many configurations of models, data, etc.
  • Masters (or equivalent research experience) in computer science (ML emphasis), or a related field
  • Proficiency with tensor-based programming languages with a focus on deep learning: Tensorflow (preferred), (py)Torch, Caffe, Theano, etc.
  • Familiarity with standard deep learning concepts and demonstrated experience with model comparison
  • Python expertise
  • Experience working with GPU accelerated environments
  • Comfort with unix and cloud computing
  • Knowledge of mathematical fundamentals: statistics, probability theory and linear algebra
  • Practical and theoretical understanding of fundamental models and algorithms in statistical learning, including comfort with modern neural-network architectures.
  • Built tools in a production software engineering environment, including use of automated regression testing, version control, and deployment systems

  • The ability to translate papers to working computational models
  • Developed algorithms and models using variational techniques
  • Experience in scientific parallel computing
  • Demonstrated via publication record or publicly available code to implement and compare many model configurations, architectures, and features
  • Written software or algorithms for high-performance computing, including SIMD or GPU performance optimization, C++, CUDA experience
  • Demonstrated experience working with >TBs of data with the requirement of just in time processing for model training
  • Experience building statistical models over multiple data types: text, images, audio, structured data, time series events, etc.
  • Linux/cloud systems admin

Freenome’s Culture
Our culture is simple. We value Empathy, Trust and Integrity. We feel from the perspectives of each other, patients and the communities we serve. We give each other the benefit of the doubt and believe we’re all working as a team towards our goals. We conduct ourselves with integrity, empowering others to grow in a collaborative environment. Freenome explicitly  prohibits any form of workplace harassment based on race, color, religion, gender, sexual orientation, gender identity or expression, national origin, age, genetic information, disability, or veteran status.

What does a successful person look like at Freenome?
You can prioritize, manage and execute your goals and align them to the mission of Freenome. You’re a steward of the culture and hold yourself and the team accountable. You believe hiring and mentorship are fundamental to building the organization. You bring ownership to your role, focus on your process and find the gaps to help Freenome reach its full potential. You welcome feedback and criticism knowing its value is to build people up and support each other, rather than tearing each other down.