Data Scientist - Deep Learning
The Data Scientist will participate in the design and prototyping of cutting-edge deep learning and statistical algorithms for analysis of genetic data.
PRIMARY RESPONSIBILITIES:
- Analyze next-generation sequencing data ranging from single research experiments to commercial data sets of millions of samples.
- Design novel deep learning architectures for application in genomics.
- Research and develop machine learning and statistical algorithms for genetic diagnostics.
- Develop Python software infrastructure to support algorithm testing and simulation studies.
- Contribute to research and product development efforts.
- Produce correct conclusions based on rigorous mathematical analysis and principles of statistics and probability.
- Produce high quality technical documentation including research reports and algorithm specifications.
QUALIFICATIONS:
- Master’s degree in engineering, applied math, statistics, or similar, PhD preferred.
- At least 2-year practical experience in scientific data analysis using software such as Python.
KNOWLEDGE, SKILLS, AND ABILITIES:
- Knowledge of deep learning techniques and theory.
- Experience applying deep learning methods (to genomic data a plus).
- Experience with using CNNs for classification and segmentation a plus
- Proficiency in at least one major deep learning framework, preferably TensorFlow.
- Strong foundation in probability theory and/or statistics including concepts like joint and conditional probability distributions, parameter estimation and hypothesis testing.
- Excellent verbal and written communication skills and the desire to work in a dynamic and collaborative environment.
- Desire to learn about human genetics and sequencing technologies.
PHYSICAL DEMANDS & WORK ENVIRONMENT:
- Duties are typically performed in an office setting.
- This position requires the ability to use a computer keyboard, communicate over the telephone and read printed material.
- Duties may require working outside normal working hours (evenings and weekends) at times.