Data Scientist ( R-00051193 )
The data science candidate is responsible for the collecting, cleaning and munging of data in ways to determine the value of the data and extract that value for discovery and decision support. Our data scientist are part detectives and business analysis. The detective work comes in handy when parsing, clustering, and stratifying the data to find and amplify signals in the data. The business analysis is used to determine what signals they should look for based their understanding of the client’s business goals for collecting the data.
The candidate will be responsible for handling the extract, transform, and load (ETL) for multi-domain data. They will be applying various machine learning algorithms to the multi-domain data to measure the performance of the algorithm and the suitability of the data. The candidates first set of algorithms will be for homomorphic encryption (HE). With those algorithms, the candidate will be experimenting with ways to increase the speed of homomorphic encryptions by orders of magnitude. The candidate must be a self-starter and honestly self-assessing to know when to seek assistance. Also, the position requires a candidate who can communicate well and able to present their work to internal and external groups.
- Bachelor's degree in Computer Science, Data Science or related field and 2-4 years of relevant experience or Masters with less than 2 years experience.
- Good understanding of machine learning algorithms, tools and platforms
- Experience in at least one of these Toolkits: numpy, scipy, scikit-learn, tensorflow, pytorch, keras, genism, vowpal wabbit, etc.
- Understanding of programming fundamentals
- Python proficiency
- Self-starter and intellectual curiosity
- Great communication skills, ability to explain predictive analytics to non-technical audience
- Proficiency in data exploration techniques and tools
- Must be able to obtain TS/SCI with CI Poly security clearance.
- Experience programming machine learning algorithms for GPUs
- Understanding of Convolutional Neural Nets
- Working knowledge of Keras
- Discernment of when and how to use machine learning regulation