Junior AI/ML Engineer

Node.Digital•Herndon, VA

5d•Hybrid

About The Position

Node.Digital is seeking a Junior AI/ML Engineer to support AI automation and machine learning solutions for government and commercial clients. The role supports fraud-related machine learning work in the IRMS environment, including data preparation, model development and evaluation, experiment tracking, model monitoring, NLP/NER pipeline integration, agent integration testing, and EPLC documentation. The position is based in Herndon, Virginia, with hybrid work.

Requirements

Bachelor’s degree in Computer Science, Data Science, Mathematics, Statistics, or a closely related field; recent graduates with strong applied ML coursework or project portfolios will be considered.
1–3 years of hands-on experience in machine learning, data science, or data engineering with Python; internships, graduate research, and project work count.
Proficiency with Python ML tools including scikit-learn, Pandas, and NumPy.
Familiarity with at least one deep learning framework, TensorFlow or PyTorch, for model evaluation and inference tasks.
Experience with ML evaluation workflows, including train/validation/test split design, cross-validation, metric computation, and documentation of results.
Experience using Jupyter notebooks for data exploration, model evaluation, and technical reporting.
Familiarity with Git-based version control and CI/CD principles, and ability to work within a structured sprint cadence with documented deliverables.
Ability to handle sensitive data responsibly and understanding of data governance, access control, and environment segregation in regulated or government settings.
Strong written communication skills and ability to produce clear, organized technical documentation for government review.

Responsibilities

Support data preprocessing and feature engineering for HRSA fraud-related datasets, including cleaning, normalization, validation, class-imbalance preparation, and train/validation/test split management.
Assist with developing, training, and evaluating supervised fraud classification models; compute and document evaluation metrics for government review in EPLC model evaluation reports.
Maintain ML experiment tracking using MLflow or an approved equivalent, documenting hyperparameters, training runs, evaluation results, and reproducibility information.
Support model drift detection and retraining pipelines by running scheduled evaluations, identifying performance degradation, and escalating findings to the AI/ML Lead Engineer and Fraud AI/ML SME.
Assist the NLP/NER pipeline team with transforming pipeline outputs into feature-compatible schemas and validating entity extraction quality against labeled references.
Develop and maintain Jupyter notebook artifacts for model exploration and reporting in EPLC deliverables, sprint reviews, and government demonstrations.
Support UiPath Maestro agent integration testing by preparing inference payloads, validating schemas, and testing integrations between ML inference APIs and the persona-based agent layer.
Implement and maintain Python data pipeline scripts for batch ingestion, feature store updates, and model scoring within the IRMS security boundary.
Follow IRMS data-handling procedures, including keeping PII/PHI within approved environments and maintaining developer/test environment segregation under HHS security policy.
Prepare supporting EPLC artifacts, including training data specifications, model evaluation appendices, data dictionary updates, and sprint retrospective documentation as directed.
Participate in code reviews and follow OWASP secure coding standards, NIST SP 800-160 engineering principles, and Node internal CI/CD quality gates.