Data Scientist - Time Series Forecasting

As a candidate for this role, you should possess the following skills and experience: - Python fluency: You should be at a daily-driver level in Python and comfortable using libraries such as pandas, numpy, scipy, and matplotlib. You should be proficient in working with notebooks and writing modular code. - Time series forecasting: You should have hands-on experience with time series forecasting methods such as ETS/Holt-Winters, ARIMA, and Croston (or similar intermittent-demand methods). Additionally, you should understand temporal cross-validation and alternative error metrics like WAPE and MASE for zero-inflated data. - Statistical intuition: You should be able to identify overfitting models and data leakage. It is essential to check for stationarity, seasonality, and structural breaks in data before model fitting. - Inventory or supply-chain math literacy: While not necessarily your primary expertise, you should have a good understanding of concepts like safety stock, reorder point, EOQ, service level/fill rate, and (s,S) policies. You should be able to interpret formulas related to these concepts. - Monte Carlo/simulation comfort: You should be able to perform simple inventory simulations using numpy and have knowledge of bootstrap, sampling distributions, and interpreting simulation results. - EDA discipline: You should have a structured approach to exploratory data analysis, starting with basic questions about data characteristics such as row count, null rate, dtype, distribution, time coverage, and key uniqueness. You should be able to create a concise summary of the dataset before proceeding with any modeling tasks. As a candidate for this role, you should possess the following skills and experience: - Python fluency: You should be at a daily-driver level in Python and comfortable using libraries such as pandas, numpy, scipy, and matplotlib. You should be proficient in working with notebooks and writing modular code. - Time series forecasting: You should have hands-on experience with time series forecasting methods such as ETS/Holt-Winters, ARIMA, and Croston (or similar intermittent-demand methods). Additionally, you should understand temporal cross-validation and alternative error metrics like WAPE and MASE for zero-inflated data. - Statistical intuition: You should be able to identify overfitting models and data leakage. It is essential to check for stationarity, seasonality, and structural breaks in data before model fitting. - Inventory or supply-chain math literacy: While not necessarily your primary expertise, you should have a good understanding of concepts like safety stock, reorder point, EOQ, service level/fill rate, and (s,S) policies. You should be able to interpret formulas related to these concepts. - Monte Carlo/simulation comfort: You should be able to perform simple inventory simulations using numpy and have knowledge of bootstrap, sampling distributions, and interpreting simulation results. - EDA discipline: You should have a structured approach to exploratory data analysis, starting with basic questions about data characteristics such as row count, null rate, dtype, distribution, time coverage, and key uniqueness. You should be able to create a concise summary of the dataset before proceeding with any modeling tasks.