Notice: While JavaScript is not essential for this website, your interaction with the content will be limited. Please turn JavasSript on for the full experience.

Senior Data Scientist (Time-Series Focused)
Torqata
Remote, NC, United States

Job Title

Senior Data Scientist (Time-Series Focused)

Job Description

Torqata, a subsidiary data and analytics company of American Tire Distributors (ATD), is looking for an expert in time series forecasting with additional knowledge in machine learning and statistics, to serve in a Senior Data Scientist role. As a domain expert in time series forecasting, you will work alongside other data scientists, data engineers, software engineers, and product managers to implement state-of-the-art forecasting services that will touch all of Torqata’s products. You will handle large and mostly structured datasets to build your models – from simple statistical models to larger global models using the latest in DNN research. Your primarily focus will be accelerating Torqata’s analytics capabilities using whatever tools and models are available. A good candidate for this role can work independently or collaboratively and has outstanding technical skills in the area of data science – classification and regression model building, diagnostics, deployment, and monitoring. You will have strong communication skills, an ability to work as part of an Agile team, and a driven curiosity to solve extraordinarily challenging problems.

Essential Duties & Responsibilities

  • Be Torqata’s domain expert in time series forecasting – hierarchical or grouped forecasting, probabilistic forecasting, augmentation of existing forecasts with exogenous data, forecasting error metrics
  • Work with large, complex data sets to build scalable ML solutions for large industry problems
  • Balance time-to-develop, time-to-market, complexity, understandability, and accuracy during development of models
  • Write production-level code in Python for components of the data science process – ingestion & transformation, feature engineering & selection, model training, validation & deployment
  • Test hypotheses using Bayesian or frequentist statistics, and explore/extract useful insights
  • Assess existing and future models for conceptual soundness, assumptions, and limitations
  • Implement the most appropriate method for extrapolating from potentially biased samples to a wider population, and providing relevant measurements
  • Build and deploy resulting models in a fashion that can be consistently monitored, and utilized across the Torqata engineering teams
  • Follow an agile development methodology

Restrictions

  • Telecommuting is OK
  • Agencies are OK

Requirements

  • BS/MS/PhD in Machine Learning, Statistics, Computer Science, Mathematics, or related subject
  • 3+ years of experience working with time series forecasting
  • Advanced understanding of machine learning in the context of business use-cases
  • 3+ years of experience with Python or R and relevant modules: SKLearn, PyTorch/TensorFlow, PyStan, Pandas, Numpy
  • Knowledge of hypothesis testing – statistical significance testing, Bayesian analysis
  • Strong analytical skills with the ability to collect, organize, analyze, and disseminate significant amounts of information with attention to detail and accuracy
  • Able to independently or collaboratively solve problems with a team of top-notch data scientists
  • Strong familiarity with SQL and Linux
  • Ability to explain complex concepts across business units
  • Professional attitude and service orientation; superb team player

Additional / Nice to Haves

  • Operational experience working with microservices and/or in a production cloud environment
  • Published papers or conference speaking engagements related to time series forecast
  • Background in supply chain and logistics work
  • Practical and theoretical knowledge of mixed models and their usage in a business context
  • Knowledge in causal inference techniques

About the Company

Torqata’s powerful data platform and suite of analytics products has been designed to enable manufacturers, retailers and distributors to work smarter, more collaboratively and drive better results across the industry.

We are a data and analytics services and software start-up in the automotive and tire industry seeking to position itself as the premier provider of such services through increased visibility throughout the tire value chain and unified reconciliation of data across disparate sources such a point-of-sale data, aggregated inventory, OE production data, product information, sales forecasts, introduction of a Blockchain ecosystem, etc.

Contact Info

Previous Senior Data Scientist, Torqata in Remote, United States Next Full Stack Python Developer, Kforce in Houston, TX, United States