Americold is looking to improve planning accuracy for appointments and labor needs at site centers by leveraging historical operational data and developing predictive planning and forecasting tools.
Requirements
Knowledge of Object-Oriented Programming Concepts, Data Structures, Data Mining, Machine Learning and AI
A strong understanding of programming languages extensively used in Data Science applications (e.g. Python, R, Scala, Java, etc.)
Modeling, scripting, and data wrangling (Pandas, Scikit-learn, etc.)
SQL proficiency in querying complex datasets.
Machine learning and statistics exposure, familiarity with time series forecasting, classification models, and regression analysis.
Cloud architecture coursework with AWS
Exposure to Apache Airflow or other workflow orchestration tools.
Responsibilities
Collaborate with data engineering and business teams to collect and understand requirements for predictive planning and forecasting tools.
Build a production-ready prototype that leverages historical operational data to improve planning accuracy for appointments and labor needs at site centers.
Extract and transform data from AWS Redshift and Oracle DB using SQL and Python-based tools (e.g. Pandas, AWS boto3, Glue).
Develop time series or classification/regression models for one or more business problems: appointment scheduling optimization and prediction, labor forecasting.
Design and implement data validation logic or anomaly detection algorithms across multi-source pipelines between on-prem and AWS.
Summarize findings into clear reports, dashboards, and/or Jupyter notebooks.
Present results with actionable recommendations to technical and operational stakeholders.
Other
Pursuing Bachelors or Masters of Science degree in Computer Science, Statistics, Engineering or equivalent
Effective analytical and problem-solving skills.
Excellent written and oral communication skills.
Demonstrated planning, task organizing and execution skills.
