Senior Engineer, Machine Learning

• Define the scope of machine learning projects and investigations; develop or provide input to schedules and budgets
• Define and control model design standards and validation criteria
• Mentor other engineers, on data ingestion, cleaning, and model development practices
• Provide oversight of model performance dashboards, ML-driven product support issues, and customer-impacting model behavior

• Design, build, and maintain end-to-end machine learning pipelines, from data preparation through training, validation, deployment, and production monitoring
• Develop, tune, and maintain predictive models, classifiers, and AI tools, such as installation quality and outage detection models, to improve accuracy and reliability
• Develop time series data mining and anomaly detection algorithms to monitor the health and performance of millions of deployed meters, sensors, radios, and IoT devices
• Recommend corrective actions and trigger field responses, or hand off findings to design and reliability engineers
• Establish testing and validation frameworks to confirm model outputs against real-world data, including coordinating field or manual verification studies
• Monitor deployed models for drift, degraded performance, or unexpected outcomes, and lead remediation efforts
• Support integration of model outputs into customer-facing dashboards and reporting tools
• Identify opportunities to improve existing infrastructure, workflows, products, and investigations with machine learning models

• Design and maintain data pipelines that integrate internal and external data sources, such as carrier network data, weather data, and manufacturing test data, into shared indices and datasets
• Interface with AWS cloud infrastructure for machine learning workloads, such as SageMaker, EC2, ECS, and Glue
• Utilize datastores such as Elasticsearch and Amazon Redshift, including indices and data structures supporting model training and inference
• Contribute to database and data architecture improvements as needed to support growing model and pipeline complexity

• Partner with engineering, digital engineering, and analytics teams to align model outputs with product and business needs
• Work closely with developers, data engineers, data scientists, and reliability, quality, and design engineering to identify and address product weaknesses revealed by fleet and field data
• Share standardized machine learning libraries, tools, and best practices across teams to reduce duplicated effort and improve consistency
• Collaborate with customer-facing teams to validate model results and translate findings into actionable insights

• Initiate and manage tickets in the team's ticketing system (e.g., JIRA)
• Write, review, validate, and audit procedures related to model development, testing, and deployment

• Bachelor's or Master's degree in Engineering, Mathematics, Statistics, Computer Science, Data Science, or Data Analytics, or equivalent practical experience

• 5+ years of related experience in machine learning, data science, or applied analytics; a Master's degree with related research may count toward a portion of this experience, depending on the topic and exposure to distributed sensor and device network data
• Strong foundation in engineering and statistical fundamentals
• Proficiency in Python and SQL; experience with NoSQL data stores, Elasticsearch, Amazon Redshift, Grafana, and Jupyter Notebooks; additional languages (e.g., C#) a plus
• Experience with version control tools and workflows (e.g., GitHub) for code and model management
• Experience with cloud-based data and machine learning platforms, preferably Amazon Web Services (e.g., S3, SageMaker, Redshift, EC2, ECS, Glue)
• Significant experience with statistics, machine learning, or other mathematical modeling and simulation techniques, including predictive modeling, classification, time series data mining, and anomaly detection
• Experience modeling performance and detecting anomalies in IoT device, sensor, or endpoint data at scale (e.g., fleets of meters, radios, or connectivity equipment)
• Experience with large-scale time series data sets and near-real-time analysis
• Experience designing, testing, validating, and deploying machine learning models in a production environment
• Experience with data visualization and business intelligence (BI) tools
• Ability to work with non-technical stakeholders to define expectations and success criteria for new models and algorithms, and to communicate results in clear, non-technical terms

• Ability to quickly develop working knowledge of metering, sensor, radio, and connectivity products
• Ability to independently solve problems and implement solutions
• Demonstrated judgment and decision-making within a defined level of authority
• Demonstrated ability to drive projects to completion
• Experience integrating external or third-party data sources (e.g., carrier network data, weather data, manufacturing/test data) into machine learning pipelines
• Experience managing machine learning pipelines end-to-end, from training through deployment and production monitoring
• Understanding of device hardware, reliability, or failure analysis sufficient to translate model findings into corrective-action recommendations (e.g., using field returns or test data)
• Familiarity with Lean Six Sigma or other continuous improvement methodologies