Software Engineer-Data Engineering, Machine Learning (ML)

American Association of Motor Vehicles•Arlington, VA

About The Position

The IT Division is responsible for the development and operations of information systems for the State and Federal agencies doing business related to or using information from the administration of motor vehicles and driver licenses. The Machine Learning (ML) Data Engineer position has core responsibilities for the design, development, deployment, and operational support of machine learning solutions on cloud infrastructure. This includes the full model lifecycle — from data acquisition and dataset preparation through feature engineering, experimentation, model training, validation, production deployment, and ongoing monitoring. Current applications include anomaly detection across high-volume messaging networks, but the scope encompasses any ML capability that strengthens system reliability, operational intelligence, and data-driven decision-making across AAMVA systems.

Requirements

Bachelor's degree in computer science, data science, statistics, mathematics, or related quantitative field. Equivalent work experience may be substituted
3–5 years of hands-on experience in data engineering, ML engineering, or applied analytics
Hands-on cloud platform experience (Azure or AWS) building and deploying data or ML solutions on managed cloud services; specific platform less important than depth of experience
Working knowledge of statistical foundations: distributions, variance, standard deviation, trend vs. seasonality, hypothesis testing, and how to apply them to real operational data
Experience with the ML experiment-to-production cycle: dataset preparation, feature engineering, model training, evaluation, and deployment
Proficiency in Python for data processing, statistical analysis, and ML model development
Strong SQL skills with understanding of relational database fundamentals: data modeling, query optimization, indexing strategies, and how SQL Server infrastructure supports production workloads (T-SQL, stored procedures, Availability Groups)
Experience building data pipelines that handle batch and streaming workloads
Experience with version control systems (Git) and CI/CD practices
Strong problem-solving skills, attention to detail, and ability to work independently on ambiguous problems
Strong written and verbal communication skills — able to explain technical findings to non-technical stakeholders and engage productively across IT, operations, and leadership; comfort operating outside the ML silo and contributing to broader technology discussions

Nice To Haves

Experience with time-series analysis, anomaly detection, or statistical process control on operational data
Familiarity with unsupervised and semi-supervised techniques (isolation forest, clustering, ensemble methods)
Experience building and managing ML model lifecycle on Azure (MLflow, Fabric ML, Azure ML) or AWS (SageMaker, Glue, Step Functions)
Familiarity with KQL (Kusto Query Language) for time-series decomposition, log analytics, or real-time data exploration
Knowledge of data modeling and dimensional modeling concepts
Experience with synthetic test data generation and model validation frameworks
Familiarity with operations and monitoring of mission-critical data platforms

Responsibilities

Designing and building dataset preparation pipelines — acquiring, cleaning, transforming, and versioning data for ML training and evaluation
Engineering features that extract meaningful signals from structured and semi-structured data sources (time-series patterns, statistical profiles, categorical encodings)
Running structured experimentation — testing multiple algorithms against defined scenarios, measuring performance, and documenting findings
Training, evaluating, and tuning ML models including regression, classification, clustering, anomaly detection, and ensemble methods
Deploying models to production on cloud infrastructure and building the pipelines that keep them running (retraining, scoring, threshold management)
Monitoring model performance in production — tracking drift, false positive rates, and detection efficacy over time
Building and maintaining batch and streaming data pipelines using Synapse, Fabric, Spark, and Event Hubs that feed ML systems
Writing and optimizing analytical queries (SQL, KQL, PySpark) for data exploration, statistical profiling, and real-time analysis
Creating validation frameworks — synthetic test data generation, backtesting against historical logs, and shadow-mode evaluation
Building dashboards and visualizations that communicate model outputs to technical and non-technical stakeholders
Collaborating with cross-functional teams to identify ML opportunities and translate operational problems into data solutions; communicating findings, trade-offs, and model behavior clearly to technical and non-technical audiences across IT, operations, and leadership