Guavus provides dynamic solutions for data-rich businesses, so that they can gain a competitive advantage in creating more meaningful customer experiences.
A Sr. Data Scientist at Guavus is responsible for creating technologies to analyze extremely large volumes of data producing actionable insights for timely decision making by our customers. A person in this role will be responsible for building machine learning models from multiple data sources in order to provide advanced analytics and reporting. The ideal candidate should have skills to quickly select the best analytics approach, perform a proof of value and write production quality code to move the research into production. The role requires exceptional coding skills along with the analytics experience to productively collaborate with our team of scientists and engineers in providing cutting edge, real-time analytics solutions to customers worldwide.
This is an exciting, technical role is based in our Montreal office. It is focused on enabling our customers to create real, practical value through advanced analytics on their streaming data. Our customers – primarily large telecom service providers – use Guavus’ analytics to increase revenues or reduce costs by better network optimization, customer experience optimization, churn management,
- Identify state of the art techniques to solve challenging predictive analytics problems
- Create machine learning pipelines, including feature engineering, to analyze terabytes of structured and semi-structured data sets
- Quickly design and conduct experiments to showcase proof of value
- Build analytics modules using Python / Scala
- Integrate analytics modules into Guavus’ real-time analytic platform and update the core machine learning functionalities in our product lines leveraging Python/ Scala.
- Work closely with other team members in converting the business problem definition into validated results with customers
- Must be self-driven and capable of prioritizing, organizing, and managing
- The ideal candidate will have a strong academic background with development experience in R/Python or Java/Scala (a big plus) with good command over data pipelining, linear algebra and statistics libraries.
- 3 -5 years of experience in analyzing large volumes of data in building predictive models is desirable
- Exposure to analytics techniques such as outlier detection, classification (RF, NB, etc.), regression, regularization, clustering, dimension reduction (PCA, etc.), reinforcement learning, stochastic methods (ARIMA, HMM, RNN, etc.), hypothesis testing will be a big plus.
- Experience with data science toolkits: pandas, scikit-learn, Jupyter, Anaconda, spark-ml, etc.
- Good understanding of calculus, linear algebra, and applied statistics
- Good understanding of CS fundamentals such as data structures and algorithms, functional programming, cluster computing, etc. and the ability to quickly translate ideas to efficient, elegant code
- Experience in high volume data scenarios and hands on knowledge of distributed systems such as Hadoop and Spark will be a big plus
Bachelor’s degree, but graduate degree preferred, in Computer Science, Engineering, Mathematics or other science field with a dissertation supported by data analysis.