Education & Experience Required:
- Bachelor”s Degree (Data Science, Analytics, Computer Science, Engineering or related field) 2-4 years experience
- Candidates with a relevant Master”s degree and PHd with 1-2 year experience
Technical Skills (Required)
- Excellent python skills (numpy/pandas/matplotlib/sklearn), version control (git), SQL and database APIs (cx_oracle, pymysql, sqlalchemy), cloud deployment experience (AWS, containerization) and familiarity with AWS services (S3/EC2/SageMaker).
- Comfortable and effective working in an Agile/Scrum environment.
- Desired qualifications: Machine Learning model development experience, familiarity with engineering/IOT systems, telematics/timeseries data analysis, and multiprocessing frameworks
(Desired)
- AWS Certified Cloud Practitioner (or higher) - desired
Soft Skills (Required)
- Excellent communication skills.
- Critical thinking and independent problem resolution.
- Ability to work both independently and on a team: collaborates and compromises effectively.
Position’s Contributions to Work Group:
– Data Scientist role to develop and deliver analytics models for client”s equipment condition monitoring for the enterprise, and for aftermarket solutions for customers and dealers.
– Condition monitoring models include failure prediction, remaining life estimation, service interval extension, and component risk models.
– Methods include machine learning, deep learning, statistical and rule-based models, and other data analytics techniques to generate actionable insights from time-series machine sensor data, other telematics data, fluids analysis, and machine inspection and service records.
Typical task breakdown:
- Collect and negotiate requirements for condition monitoring models.
- Access, analyze, cleanse, preprocess data; cultivate ground truth data sets as needed.
- Explore and downselect modeling approaches.
- Develop, backtest and validate models in python, in both on-prem and AWS environments.
- Compute model performance metrics.
- Participate in daily standups and sprint reveals in Scrum/Agile environment.
- Explain and defend modeling decisions and performance results.
- Support model in production, resolve incident tickets.
Interaction with team:
– Daily standup with team
– Technical touchpoints with the lead
Work environment:
– Hybrid work model