Senior AI/ML Engineer with Security Clearance

About the role G3 is looking for a Senior AI/ML Engineer to help design, develop, and deploy machine learning capabilities for sensor data in direct support of national security missions. This is a technically demanding role. We are not looking for someone who has applied pre-built models to standard datasets. We need someone who understands how modern ML architectures work, can make principled decisions about which approaches fit a given problem, and can implement them correctly from the ground up in a mission context.

The work centers on using AI and ML to detect, classify, and characterize events and phenomena in sensor data, turning complex phenomenology into automated, scalable analytic capabilities. You will work directly with scientists, algorithm engineers, and intelligence analysts to understand the problem, design the right solution, and deliver operational-prototype code that runs reliably in classified R&D environments. You will also be a technical anchor for junior developers: setting standards, reviewing work, and helping others grow. Responsibilities

• Design and implement ML models that detect, classify, and characterize events and signatures in sensor data. Bring enough architectural depth to select and tune the right approach rather than defaulting to the nearest framework.
• Build end-to-end ML workflows: training data preparation, model development, evaluation, optimization, and deployment into operational or near-operational environments.
• Work closely with scientists and domain experts to understand sensor phenomenology, data characteristics, and mission requirements, and turn that knowledge into sound ML design decisions.
• Write production-quality Python that is modular, well documented, testable, and maintainable by others, not scripts that only work in a notebook.
• Evaluate model performance rigorously against operationally meaningful metrics. Identify failure modes, characterize edge cases, and iterate systematically.
• Integrate ML capabilities into existing analytic pipelines and software environments, and own that integration through deployment.
• Mentor junior AI/ML developers: review code, set team-wide standards, find gaps in understanding, and actively build the technical skills of those around you.
• Contribute to trade studies and architecture decisions with technically grounded recommendations on ML approaches, tooling, and implementation.
• Prepare and deliver technical briefings and written documentation to customer and program stakeholders, clear enough to hold up to scrutiny. Required qualifications
• Active Top Secret clearance, with the ability to obtain SCI and a CI polygraph.
• Bachelor's degree in Computer Science, Electrical Engineering, Physics, Mathematics, Data Science, or a closely related technical field. Advanced degree strongly preferred.
• 7+ years of hands-on ML experience, with a track record of developing and deploying models in real operational or production environments, not just research prototypes.
• Deep working knowledge of modern ML architectures (CNNs, transformers, object detection frameworks, and related methods), including why they work, where they break, and how to adapt them to non-standard data.
• Expert-level Python, including PyTorch, TensorFlow, or equivalent frameworks at production scale.
• Experience building complete ML pipelines: data ingestion and preprocessing, annotation and labeling strategy, training, evaluation, and deployment.
• Demonstrated ability to work directly with domain scientists or subject matter experts, translating their requirements into ML design and ML outputs back into domain terms.
• Hands-on experience mentoring or leading junior developers through code review, standard-setting, and knowledge transfer, not just oversight.
• Strong documentation and communication skills, including writing technical reports and briefing findings to non-ML audiences. Preferred qualifications
• Experience applying ML to sensor data such as EO/IR, hyperspectral, SAR, or other remote sensing data. Familiarity with the phenomenology matters here, not just the algorithms.
• Background in computer vision, image classification, anomaly detection, or event characterization with scientific or sensor data.
• Experience training models on simulated or synthetic data, including how to manage the gap between simulated training data and real operational data.
• Familiarity with MLOps practices: experiment tracking, model versioning, reproducible pipelines, and deployment in constrained or classified environments.
• Experience with large-scale geospatial or scientific datasets, and comfort with noisy, sparse, or domain-specific data.
• Proficiency with scientific Python libraries such as NumPy, SciPy, pandas, OpenCV, or scikit-learn.
• Experience with cloud ML training environments (AWS, Azure, or equivalent) and GPU-accelerated compute.
• Prior experience supporting defense, intelligence, or classified customers.
• Ph. D. in a relevant quantitative field such as Astronomy, Physics, Computer Science, or Electrical Engineering, especially where the doctoral work applied ML to real scientific data.