AI Engineer (Remote)

GovStar develops technology and creative solutions to support government and national security objectives. The company is seeking a Senior AI Engineer to design, evaluate, secure, and operationalize large-language-model services within approved government cloud environments, while integrating data sources and communicating technical risks and recommendations to stakeholders.

Responsibilities:

• Design and implement AI-enabled services using large language models and approved enterprise AI frameworks
• Develop retrieval-augmented generation (RAG), prompting, tool-use, orchestration, classification, extraction, summarization, and other model-backed workflows
• Integrate AI services with approved structured and unstructured data sources in the centralized data platform
• Develop evaluation frameworks for accuracy, relevance, groundedness, robustness, safety, and mission usefulness
• Implement guardrails and controls to reduce hallucination, inappropriate disclosure, prompt-injection risk, data leakage, and other AI-specific failure modes
• Design model governance, traceability, auditing, and human-in-the-loop patterns appropriate for sensitive government use cases
• Support AI capabilities such as investigative/hotline analysis, natural-language querying, trend and anomaly detection, link analysis, pattern recognition, automated tagging/classification, drafting, eDiscovery support, translation, and transcription
• Deploy and integrate solutions using approved services such as Azure OpenAI, OpenAI models, commercial LLM providers, and other enterprise AI frameworks where permitted
• Partner with platform and cloud engineers to productionize AI services using secure CI/CD, IaC, containerization, monitoring, and scalable deployment patterns
• Support pilot design, data-quality evaluation, AI-output validation, migration of existing AI tools, and transition from pilot to sustained operations
• Clearly communicate model limitations, technical tradeoffs, evaluation findings, risks, and recommendations to technical and non-technical stakeholders