Data Scientist - Reliability
The Reliability Engineering team is looking for a Data Scientist who is interested in leveraging statistical and computational tools to improve how Tesla products are designed. Our team’s goal is to predict future vehicle risks based on existing field data. Models and analyses we build will be used in two ways: to isolate high risk vehicles in the current fleet, and to inform the design and engineering communities on how future vehicles should be built. Both of these with the intention of delighting our customers with exceptional vehicle reliability.
Responsibilities:
- Contribute to the development of our data pipelines by implementing novel algorithms for retrieving, analyzing and visualizing data
- Apply statistical analysis on vehicle data to drive decision making in reliability
- Extract useful statistics and usage profiles from the existing fielded fleet in order to drive accurate design requirements for next generation vehicles
- Apply machine learning algorithms to predict degradation trends and create predictive maintenance schemes for the vehicle fleet
- Create visualization methods to communicate data in a meaningful and actionable manner
Requirements:
- Advanced knowledge of Python. Working knowledge of pandas, scipy, numpy, IPython
- General knowledge of data structures, architectures and languages such as SQL
- Demonstrated ability to build robust data infrastructures
- Experience and interest in data visualization techniques. Ability to convey complex analyses with intuitive visual methods while also effectively communicate findings
- Bachelor of Science in computer science, engineering, statistics or related discipline. MS or PhD preferred
- Minimum 1 year of experience in a data analysis position or equivalent experience
- Knowledge of applied statistics, including multivariate statistical analysis and time series analysis
- Working knowledge of Hadoop ecosystem, in particular one of more of the following: Hive, Spark, Impala
- Strong experience applying quantitative methods and machine learning to large-scale engineering problems. Working knowledge of scikit-learn, statsmodels and other machine learning/ statistics/ data mining packages
- Experience with reliability analytical methods (e.g. Weibull analysis, reliability growth analysis, reliability block diagrams etc.)