Machine learning engineer Jobs in Salt Lake City, UT

The ideal candidate will leverage their expertise in machine learning to design and implement scalable models and processes for a variety of classification and data analysis tasks. A key focus of this role will be the classification of powersports, marine, and RV vehicles based on VIN (Vehicle Identification Number), HIN (Hull Identification Number), or PID (Product Identification). Additionally, the role will involve developing datasets and algorithms to support vehicle and part classification while contributing to broader machine learning initiatives.

What you'll do :

• Design, develop, and deploy machine learning models for diverse classification tasks, including parts, VIN, HIN, and PID-based vehicle identification.
• Implement scalable algorithms and solutions to enhance the accuracy and efficiency of classification systems.
• Build and maintain datasets and feature engineering to support classification models and metadata generation.
• Ensure datasets are accurate, comprehensive, and updated regularly.
• Explore and incorporate new data sources to improve model performance and extend metadata for vehicle and part identification.
• Create ETL (Extract, Transform, Load) pipelines to ingest, clean, and preprocess data from various sources.
• Automate data workflows to streamline the integration of new data sources and enhance scalability.
• Collaborate with internal teams, OEMs, and aftermarket suppliers to acquire and integrate relevant data.
• Work with cross-functional teams to align machine learning projects with organizational objectives.
• Continuously evaluate and fine-tune machine learning models to improve their performance and reliability.
• Conduct experiments and validation tests to ensure the robustness of implemented solutions.
• Document workflows, methodologies, and results for internal use and knowledge sharing.
• Develop reports and visualizations to communicate findings and model performance to stakeholders.

Requirements

What you should have :