Principal AI Engineer - Blue Ring

BLUE ORIGIN•Reston, VA

1d•$230,773 - $323,082•Onsite

About The Position

Blue Origin is seeking a Principal AI Engineer to join the Blue Ring software team within Blue National Security. The role will lead development of AI and machine learning capabilities for Blue Ring’s ground system, supporting satellite operations at fleet scale. These capabilities include multi-agent monitoring, conversational knowledge retrieval, and ML-driven anomaly detection. The engineer will build agent tooling, maintain retrieval-augmented generation (RAG) knowledge pipelines, develop telemetry ML models, and support production AI services used by operators. The position contributes to the AI/ML systems that form the intelligence layer for Blue Origin’s satellite fleet operations. Work includes developing and integrating multi-agent components, building retrieval and answer-generation workflows, supporting human-in-the-loop decisions, deploying containerized services, and preparing telemetry data for model training and evaluation. The role is onsite at an office in Kent, Washington; Renton, Washington; or Reston, Virginia.

Requirements

Bachelor’s degree in Computer Science, Machine Learning, Electrical Engineering, or a related field.
10 or more years of professional software development experience with exposure to AI/ML applications.
Proficiency in Python and familiarity with at least one AI/ML framework, such as PyTorch or TensorFlow.
Experience building or contributing to LLM-based applications, RAG pipelines, or agent-based systems.
Exposure to time-series data processing, anomaly detection, or ML model training on sensor or telemetry data.
Experience with cloud platforms (AWS preferred), containerization (Docker), and basic CI/CD practices.
Solid understanding of computer science fundamentals, including data structures, algorithms, object-oriented design, and version control.
Strong analytical and problem-solving skills with attention to detail.
Good written and verbal communication skills for documentation and team collaboration.
Able to work onsite in the Kent, WA, Renton, WA, or Reston, VA office.
Must meet the stated export-control eligibility criteria: U.S. citizen or national, U.S. permanent resident, or lawfully admitted into the U.S. as a refugee or granted asylum.

Nice To Haves

PhD in Computer Science, Machine Learning, Electrical Engineering, or a related field.
Experience with AWS Bedrock, SageMaker, or equivalent managed AI/ML services.
Experience with Databricks or similar platforms for data engineering and ML workflows.
Familiarity with vector databases such as Pinecone, pgvector, or FAISS, and embedding model pipelines.
Experience with time-series ML architectures such as RNNs, LSTMs, or Transformers for forecasting or anomaly detection.
Exposure to multi-agent system frameworks or agentic AI orchestration patterns.
Familiarity with spacecraft operations, satellite telemetry, or aerospace engineering concepts.
Experience with infrastructure-as-code tools such as Terraform and Kubernetes.
Experience with real-time data streaming systems such as Kafka or Kinesis.
Familiarity with reinforcement learning concepts or graph neural network applications.
Interest or experience in model compression, edge deployment, or resource-constrained inference.

Responsibilities

Implement and test multi-agent components for telemetry analysis, anomaly detection, and reporting.
Build and maintain integrations connecting agents to ground system services.
Develop agent tooling, prompt templates, and orchestration logic for LLM service integrations.
Build and maintain RAG pipeline components, including document ingestion, embedding generation, and vector database indexing.
Implement retrieval strategies and answer-generation workflows for operator queries.
Implement components for human-in-the-loop decision workflows, including investigation packages and remediation options.
Deploy and maintain containerized AI services using Docker and AWS infrastructure.
Contribute to audit trails, observability systems, and production operations, including incident response.
Prepare and transform spacecraft telemetry data for ML model training.
Assist in training and evaluating anomaly-detection and time-series forecasting models using telemetry data.
Support automated model retraining pipelines incorporating new flight data and operator feedback.
Develop model evaluation scripts and metrics tracking.

Benefits

Medical, dental, and vision insurance.
Basic and supplemental life insurance.
Paid parental leave.
Short- and long-term disability.
401(k) with company match of up to 5%.
Education Support Program.
Stock options for regular employees working at least 20 hours per week.
Paid time off of up to four weeks per year based on weekly scheduled hours.
Up to 14 company-paid holidays.
Depending on role type and job level, employees may be eligible for benefits and bonuses based on individual contributions or other factors; amounts and eligibility are not guaranteed.