AI/Machine Learning Engineer II

At Niagara, we’re looking for Team Members who want to be part of achieving our mission to provide our customers the highest quality most affordable bottled water.

Consider applying here, if you want to:   

• Work in an entrepreneurial and dynamic environment with a chance to make an impact.   
• Develop lasting relationships with great people.   
• Have the opportunity to build a satisfying career.

We offer competitive compensation and benefits packages for our Team Members.

AI/Machine Learning Engineer II

As an AI/Machine Learning Engineer, you'll work on training, evaluating, and serving large AI models, internet-scale dataset building, and prototype new research and product ideas. Your responsibilities include pre-training, optimizing inference throughput, continual learning, implementing interfaces, designing, testing, and optimizing new neural net architectures, and internet-scale data scraping. Additionally, you will work with Product Management and data scientists to build and constantly lead excellence in our products.
 

Detailed Description

• Develop & Design Predictive analytic systems through Continuous Monitoring of critical asset parameters
  • Data Analysis: Conduct in-depth analysis of large and complex datasets related to equipment sensor data, maintenance logs, and other relevant sources using machine learning programs and libraries such as Python with libraries like NumPy, Pandas, and SciPy.
  • Model Development: Design and implement advanced predictive maintenance models using machine learning algorithms and techniques from libraries such as scikit-learn, TensorFlow, or PyTorch. Apply regression, classification, clustering, and time series analysis algorithms to develop accurate predictive models.
  • Feature Engineering: Utilize machine learning libraries to extract, transform, and engineer relevant features from raw data. Use feature selection techniques and data preprocessing methods available in libraries like scikit-learn to optimize model performance.
  • Model Training and Evaluation: Utilize machine learning frameworks to train and fine-tune predictive maintenance models using historical data. Evaluate model performance using metrics such as accuracy, precision, recall, and F1-score, leveraging libraries like scikit-learn or Keras.
  • Data Visualization and Reporting: Utilize data visualization libraries like Matplotlib or Plotly to create interactive visualizations and reports that effectively communicate insights derived from predictive maintenance models.
  • Collaboration and Cross-functional Communication: Collaborate with maintenance engineers, data engineers, domain experts, and stakeholders from different departments, using tools like Jupyter Notebooks or Git, to share code, insights, and results. Foster effective communication and knowledge sharing within the team.
  • Data Governance and Quality Assurance: Ensure data integrity, quality, and security throughout the predictive maintenance process. Implement data cleansing, validation, and quality control measures using libraries like Pandas or PySpark to ensure accurate and reliable model outputs.
  • Research and Innovation: Stay updated with the latest advancements in machine learning and predictive maintenance. Explore new machine learning algorithms, libraries, and techniques that can enhance predictive maintenance capabilities.
  • Documentation: Maintain clear and concise documentation of methodologies, code implementations, and model specifications using tools like Markdown or Sphinx. Document experiments, findings, and best practices for future reference and knowledge sharing.
  • Continuous Improvement: Continuously explore ways to improve model performance, scalability, and efficiency using machine learning libraries and frameworks. Keep up with the latest research papers and developments to incorporate cutting-edge techniques into predictive maintenance models.
  • Training and Knowledge Transfer: Share expertise and insights with colleagues, stakeholders, and other team members. Conduct training sessions or workshops to promote understanding and effective utilization of machine learning programs and libraries.

• Develop and design advanced systems for asset reliability for all manufacturing assets across all Niagara plants.
  • Machine Learning Pipeline Development (ML Ops): Design, develop, and implement end-to-end machine learning pipelines, from data ingestion and preprocessing to model training, evaluation, and deployment. Implement automation and orchestration techniques to ensure reproducibility and scalability.
  • Data Analysis and Modeling: Apply advanced statistical