Durability Data Analytics Engineer

Rivian•Irvine, CA

6d•Remote

About The Position

The Durability Attribute engineers predict and test the range of loads and stresses Rivian vehicles will experience over life and need to be durable for—a challenging prediction problem with many unknowns. With access to large-scale vehicle fleet data and software-defined vehicles that offer greater scope for control and autonomy, vehicle durability is at the forefront of data-driven engineering at Rivian ensuring the “Forever” in Rivian’s mission of Keeping the World Adventurous Forever. As a Durability Data Analytics Engineer, you will be part of the collaborative team of simulation, test, and data analytics engineers making Rivian vehicles adventurous yet durable for our customers. You will work with real-world, virtual, and vehicle test data, to identify, classify, and predict customer travel, driving, road surfaces, and other use cases, as well as undesirable events such as ground strikes or recovery events that impact loads on the vehicle and cause lifetime degradation. You will also develop and correlate a hybrid of data- and physics-based models to predict loads and vibrations at various suspension, body, and propulsion components to reduce the need for repeated complex-physics simulations and instrumented vehicle tests. Through your work, you will have the opportunity to improve process automation, efficiency, and standardization of vehicle durability target development.

Requirements

Bachelor's degree in Mechanical Engineering, Aerospace Engineering, or Computer Science.
A good understanding of vehicle physics and automotive engineering.
Passion for Rivian’s mission and commitment to fostering an inclusive, respectful, and collaborative workplace.
1-3 years experience in data analytics for automotive or related engineering applications.
Experience with Python (Pandas, Numpy), PySpark, Matlab, and SQL applied to big datasets.
Demonstrated expertise with various machine learning techniques applied to automotive use data, test data, or simulation data.
Experience in fatigue life estimation (e.g., Rainflow counting, Miner’s Rule) and statistical methods (e.g., Weibull analysis)
Demonstrated experience in analytical writing (reports, white papers, publications) and clear-and-concise presentations, with attention to detail.

Nice To Haves

Project or research experience with large time series datasets is desired.

Responsibilities

Develop fleet data analyses mapping travel, driving, road infrastructure, and vehicle usage across the US and other markets.
Build classification models to identify usage archetypes, usage frequency, and their correlations.
Develop and validate machine learning models to estimate vehicle loads, drive events, fatigue damage, and vibration spectra from vehicle CAN data signals, wheel-force transducer data, or accelerometer data.
Develop models to standardize vibration spectra at different vehicle locations across vehicle variants and platforms.
Learn from simulation and test data to develop a library of regression models for variance in component loads and fatigue damage with variations in system design parameters and usage types.
Automate processes such as component block cycle development and accelerated duty cycle development from real-world duty cycle.
Apply machine learning on past CAE results to develop ways to reduce the number of CAE test cases and runs.