Lead Hospitality Demand Forecasting for New Venues

Halcyon Knights; Melbourne; On-site; AUD 120,000 - 190,000; Full time

Job summary

Halcyon Knights is seeking a data scientist to own a demand forecasting model for recently acquired hospitality venues across Australia. The role involves designing a model for revenue, covers and product demand at daily and intraday levels, and addressing cold-start challenges using data from comparable venues.

You will bring in external drivers like weather and events, integrate disparate data sources, and deploy to production with automated retraining and monitoring.

Qualifications

• 5+ years in data science with demand or sales forecasting models in production.
• Proficient in Python and SQL.
• Experience with forecasting methods such as LightGBM, XGBoost, Prophet or ARIMA.

Responsibilities

• Design and build a demand forecasting model for recently acquired venues, covering revenue, covers and product demand at daily and intraday levels.
• Address cold-start by leveraging data from comparable venues and hierarchical approaches.
• Incorporate external drivers like weather, holidays, events and promotions.
• Clean, align and integrate venue data from diverse POS systems.
• Deploy the model to production with automated retraining, monitoring and drift detection.
• Set and report accuracy metrics by venue, region and venue type.
• Collaborate with operations, finance and venue managers to ensure forecasts are trusted and used.
• Document the approach for future acquisitions.

Skills

Python

SQL

Forecasting

Time series

Cloud deployment

Education

Tertiary qualification in data science/statistics/engineering

Tools

LightGBM

XGBoost

Prophet

ARIMA

Azure

AWS

Databricks

PySpark

MLflow

Job description

Halcyon Knights is seeking a data scientist to own a demand forecasting model for recently acquired hospitality venues across Australia. The role involves designing a model for revenue, covers and product demand at daily and intraday levels, and addressing cold-start challenges using data from comparable venues.

You will bring in external drivers like weather and events, integrate disparate data sources, and deploy to production with automated retraining and monitoring.