Reinforcement Learning AI Engineer

Booz Allen Hamilton • McLean, VA

5d • $99,000 - $225,000

About The Position

The Reinforcement Learning AI Engineer develops, trains, tests, deploys, and maintains AI models that learn from data to address real-world, mission-critical problems. The role translates reinforcement learning research into operational capabilities and production-grade code, collaborating with machine learning engineers and professionals across space, military, intelligence, engineering, and AI domains.

The engineer designs and trains reinforcement learning and multi-agent reinforcement learning algorithms for complex decision-making, develops scalable training pipelines, and builds and evaluates agents in simulated environments. The position applies a range of reinforcement learning techniques, works with domain experts to align reward structures and evaluation metrics with mission objectives, and implements distributed training workflows. It also transitions trained models into production systems, contributes to system architecture and performance optimization, and may extend high-performance components into C++ or Rust.

Requirements

• At least 1 year of experience developing and training reinforcement learning agents.
• Experience in AI, data science, machine learning engineering, or software engineering.
• Experience with Gym or PettingZoo interfaces.
• Experience with machine learning frameworks such as PyTorch, TensorFlow, or JAX.
• Experience developing technical solutions using Rust, Python, or C++.
• Knowledge of reinforcement learning and artificial neural networks.
• Secret clearance required.
• Bachelor's degree in a STEM field.

Nice To Haves

• Experience applying reinforcement learning to autonomy, control systems, or mission-scale problems.
• Experience with multi-agent reinforcement learning.
• Experience with AFSIM or other high-fidelity simulation environments.
• Experience with embedded systems programming in Rust, C, or C++.
• Experience in GPU programming, including CUDA or RAPID.
• Experience developing in-space solutions.
• Knowledge of modern software design patterns, including microservice design and orchestration in Kubernetes deployment.
• TS/SCI clearance.
• Master’s degree in computer science, AI, engineering, or a related field.

Responsibilities

• Design, implement, and train reinforcement learning and multi-agent reinforcement learning algorithms for complex decision-making problems.
• Develop scalable training pipelines using Python and modern machine learning frameworks.
• Build and evaluate agents in Gym or PettingZoo, high-fidelity simulators, or custom simulated environments.
• Apply policy optimization, value-based learning, model-based reinforcement learning, and imitation learning techniques.
• Collaborate with domain experts to define reward structures, constraints, and evaluation metrics aligned with mission objectives.
• Implement distributed training workflows using cloud compute, containerization, and orchestration technologies.
• Transition trained models into production systems using strong software engineering practices.
• Contribute to system architecture and performance optimization in Python, with opportunities to develop high-performance components in C++ or Rust.

Benefits

• Health, life, and disability benefits.
• Financial and retirement benefits.
• Paid leave.
• Professional development.
• Tuition assistance.
• Work-life programs.
• Dependent care.
• Recognition awards for exceptional performance and demonstration of company values.

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level