Machine Learning Engineer - Mid

SOSi•Huntsville, AL

3d•Onsite

About The Position

SOSi is seeking a Mid-Level Machine Learning Engineer to support the development, integration, and deployment of machine learning capabilities for a government customer in Huntsville, Alabama. The selected candidate will contribute to an MLOps environment supporting national security missions and large-scale analytical workloads, helping transform machine learning prototypes into reliable, maintainable, production-ready capabilities. The engineer will work alongside data scientists, data engineers, and software developers in an Agile environment to deliver solutions aligned with mission requirements. The work spans the machine learning lifecycle, including data preparation, experimentation, evaluation, deployment, monitoring, and continuous improvement. Technologies used by the team include Python, PyTorch, scikit-learn, MLflow, PostgreSQL, Docker, Linux, GitLab, Jupyter Notebooks, and supporting MLOps infrastructure. This is an office-based position in Huntsville, Alabama, involving mission-sensitive data and secure computing environments. The role includes collaboration with multidisciplinary engineering and government stakeholder teams. Occasional remote work or travel may be available based on program requirements.

Requirements

Active Top Secret clearance with SCI eligibility.
Bachelor’s degree in Computer Science, Engineering, Mathematics, Statistics, Data Science, or a related quantitative discipline.
3–5 years of relevant experience in machine learning engineering, software development, or applied data science.
Proficiency in Python and common machine learning libraries.
Practical experience training, evaluating, and deploying machine learning models.
Experience with machine learning frameworks such as PyTorch, TensorFlow, or scikit-learn.
Experience using SQL and relational databases.
Working knowledge of Linux development environments and command-line tools.
Experience with Git, collaborative development, and pull request workflows.
Familiarity with Docker and containerized application deployment.
Understanding of automated testing, including unit and integration testing.
Knowledge of REST APIs and integration between software components.
Understanding of CI/CD principles and Agile development methodologies.
Ability to independently implement, troubleshoot, and deliver assigned engineering tasks.

Nice To Haves

Experience deploying and maintaining machine learning models in production.
Experience with MLflow or similar experiment tracking and model registry platforms.
Familiarity with Triton Inference Server or comparable inference serving platforms.
Experience developing computer vision, NLP, classification, or anomaly detection solutions.
Experience integrating machine learning models into ETL/ELT pipelines.
Familiarity with Apache Airflow or similar workflow orchestration tools.
Experience with PostgreSQL, large-scale datasets, or distributed processing.
Understanding of model monitoring, data drift, reproducibility, and rollback strategies.
Familiarity with GitLab CI/CD, automated validation, and container registries.
Experience developing solutions in restricted, offline, or on-premises computing environments.
Experience supporting DoD or Intelligence Community missions.
Exposure to high-performance computing and multi-INT analytical workloads.

Responsibilities

Develop, train, evaluate, and improve machine learning models for operational applications.
Apply supervised and unsupervised learning techniques to structured and unstructured datasets.
Perform feature engineering, data preprocessing, and dataset validation.
Select algorithms and evaluation metrics based on project requirements; conduct experiments, compare model performance, and document technical findings.
Identify and address model performance issues, including overfitting, data leakage, and prediction errors.
Convert research prototypes and notebooks into maintainable Python modules and services.
Integrate trained models into existing applications, APIs, and processing workflows.
Package and deploy model inference services using containerized environments.
Implement consistent model loading, configuration management, and error handling.
Support model versioning, experiment tracking, and reproducible deployments; investigate and resolve inference, integration, and performance issues.
Contribute to automated model validation and deployment workflows.
Write clean, modular, testable, and maintainable Python code.
Develop unit tests for preprocessing, feature engineering, and model-related functionality; develop integration tests for model interactions with dependent systems and smoke tests for deployed services.
Participate in peer code reviews, address identified defects, and use Git-based development workflows and established CI/CD practices.
Maintain technical documentation covering implementation, testing, and operational procedures.
Collaborate with data engineers to establish reliable training and inference datasets.
Implement dataset quality, schema consistency, and missing-value validation checks; work with PostgreSQL and other approved data sources.
Use MLflow or comparable tools to manage experiments and model artifacts; support model registry workflows, version promotion, and rollback procedures.
Monitor model behavior, resource utilization, and inference performance; identify data drift, performance degradation, and operational anomalies.
Support reproducible model training and deployment processes.
Complete assigned development tasks within agreed sprint objectives, communicate progress, blockers, and delivery risks, and provide realistic estimates.
Coordinate interface requirements with data engineering and software development teams.
Follow architecture, coding, testing, and documentation standards; identify proposed scope changes before implementing unapproved functionality.
Participate in technical discussions, planning sessions, and code reviews.