
Wireless and Edge AI/ML Engineer
Booz Allen Hamilton3 days ago
Annapolis Junction, MD, USAMid Level
H1B sponsor
Base Salary
$99k - $225k/yr
Responsibilities
- Design, train, and evaluate ML models for wireless network use cases using network telemetry, KPIs, and RF data.
- Build, label, version, and maintain reproducible training datasets from network performance counters, logs, packet captures, and I/Q samples.
- Optimize models for real-time GPU inference through quantization, pruning, TensorRT or ONNX Runtime conversion, CUDA profiling, and latency and throughput benchmarking.
- Deploy containerized model inference services on Kubernetes using NVIDIA Triton Inference Server, NVIDIA GPU Operator, and MIG or time-sliced GPU sharing.
- Develop and validate AI-driven O-RAN xApps and rApps that consume network telemetry and return control decisions.
- Implement MLOps capabilities including experiment tracking, model registries, retraining triggers, drift monitoring, and CI/CD for model deployment.
- Benchmark resource contention between wireless signal-processing and AI inference workloads and document the operating envelope for co-location.
- Apply ML security and assurance practices including model provenance, dataset lineage, adversarial robustness testing, and DoD AI assurance guidance.
- Create model cards, training and deployment procedures, benchmark reports, and reproducibility guides.
- Collaborate with network, RF, and cybersecurity engineers and brief technical results to government stakeholders.
Requirements
- At least 3 years of experience training and evaluating ML models with frameworks such as PyTorch or TensorFlow and exporting models to deployment formats such as ORAN or ONNX.
- At least 3 years of Python software development experience, including production-quality code, testing, and version control practices.
- Experience deploying ML models as services in Kubernetes.
- Experience testing and integrating O-RAN systems in virtual machines.
- Bachelor's degree in a STEM field.
- Ability to obtain a Secret clearance.
- Preferred experience includes RF theory, digital signal processing, physical-layer algorithms, GNU Radio, MATLAB, AI-RAN, NVIDIA Aerial, GPU-accelerated Layer 1, O-RAN RIC architectures, xApps, rApps, E2, A1, 5G/LTE KPIs, and 3GPP performance-management data.
- Preferred experience includes LLM inference optimization, vLLM, NVIDIA TensorRT-LLM, transformers, graph neural networks, reinforcement learning, C/C++, systems programming, GPU-accelerated computing, MLflow, Kubeflow, DVC, DoD AI assurance, responsible AI guidance, DoD STIG, and RMF requirements.
- Ability to work hands-on in a fast-paced lab environment, produce clear technical documentation, brief technical and non-technical audiences, and resolve complex problems with limited direction.
Benefits
- Health, life, disability, financial, and retirement benefits.
- Paid leave, professional development, tuition assistance, work-life programs, and dependent care.
- Recognition awards for exceptional performance and demonstration of company values.
- Full-time and part-time employees working at least 20 hours per week on a regular basis are eligible for Booz Allen benefit programs, with select offerings for employees below that threshold.
- The role includes hands-on work in a lab and may require work at a Booz Allen or customer facility depending on the position's remote, hybrid, or onsite designation.
Categories
Data EngineeringML Engineering
About Booz Allen Hamilton
Booz Allen Hamilton builds and integrates software, analytics, AI, cloud, and cybersecurity solutions while providing engineering and consulting services, primarily for U.S. defense, intelligence, and civil government clients, with select commercial partnerships. Its business is largely government contracting and professional services delivering mission systems and cyber defense. Founded in 1914 and headquartered in McLean, Virginia, Booz Allen is a public company traded on the NYSE (ticker: BAH).