Amazon Applied Scientist Demand Forecasting SCOT 

United States, New York, New York 

796179786

Key job responsibilities
Design new deep learning algorithms and creative approaches to push time series forecasting performance beyond the current state-of-the-art.
A day in the life
You will deep dive into massive Amazon time series datasets, leveraging Python and PySpark to conduct exploratory data analysis. You will prep the data pipeline - writing Python code to train, evaluate, and debug the latest deep learning models. You will design novel neural network architectures, brainstorming with fellow scientists and engineers on ways to push the boundaries of foundation time series modeling. You may explore and benchmark open-source solutions, identifying opportunities to enhance Amazon’s forecasting capabilities. You will document your discoveries and prepare them for internal knowledge-sharing sessions or external conferences, ensuring that your innovations help shape the future of demand forecasting - both within Amazon and the broader research community.

- 3+ years of building models for business application experience
- PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
- Experience programming in Java, C++, Python or related language
- Experience in any of the following areas: algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, high-performance computing
- Experience in state-of-the-art deep learning models architecture design and deep learning training and optimization and model pruning

- Experience using Unix/Linux
- Experience in professional software development