Senior AI Engineer

Expression•Shirlington, VA

12h•$180,000 - $210,000

About The Position

Expression is seeking experienced Senior AI Engineers to design, develop, integrate, and evaluate advanced artificial intelligence and large language model capabilities supporting the Department of Homeland Security. The role supports implementation of secure, mission-focused AI applications within the DHS AI environment, including multi-agent AI workflows, retrieval-augmented generation, intelligent document processing, and AI evaluation frameworks for investigative analysis, information retrieval, report generation, and human decision support. The engineers will bring hands-on expertise in one or more areas: multi-agent orchestration and model integration; RAG and AI-powered application development; or AI assurance, evaluation, and observability. They will collaborate with AI architects, data engineers, and Government stakeholders to build, test, and improve production-quality AI capabilities within an established enterprise environment. The position is located in Shirlington, Virginia and requires a Secret clearance with eligibility for TS/SCI.

Requirements

Bachelor's degree in Computer Science, Software Engineering, Computer Engineering, Data Science, Artificial Intelligence, or a related technical field, or equivalent relevant professional experience.
At least 9 years of professional experience in software engineering, AI engineering, or AI test automation, including substantial recent hands-on experience developing, integrating, or evaluating enterprise AI applications.
Demonstrated experience building, deploying, or evaluating LLM-powered and generative AI solutions in enterprise or production-oriented environments, with direct implementation responsibility.
Strong Python programming and debugging skills, including backend development, RESTful APIs, asynchronous programming, modular application design, and automated testing; experience with FastAPI, Pydantic, asyncio, or comparable technologies.
Hands-on expertise in at least one specialization: multi-agent orchestration and model integration; grounded AI applications and RAG; or AI assurance, evaluation, and observability.
For multi-agent orchestration: experience developing applications with LangGraph, LangChain, Semantic Kernel, AutoGen, or comparable frameworks, and implementing model/tool integrations, structured agent handoffs, stateful workflows, prompt management, and bounded execution.
For grounded AI applications and RAG: experience designing and implementing RAG pipelines, document processing, source-grounded generation, prompt/output validation, citation handling, and AI-assisted document generation or human-reviewed workflows.
For AI assurance, evaluation, and observability: experience developing automated LLM evaluation frameworks, regression suites, model reliability tests, adversarial testing, prompt-injection defenses, and application observability.
Experience integrating AI applications with enterprise data platforms, APIs, or retrieval services, including secure access to structured or unstructured data.
Experience designing and implementing secure AI applications, including authentication, authorization, permission-aware data access, sensitive-data handling, and appropriate human oversight.
Experience with LLM application evaluation and reliability, including response quality, factual grounding, hallucination mitigation, error handling, and model performance measurement.
Experience with modern software engineering practices, including Git, code reviews, CI/CD pipelines, automated testing, and software configuration management.
Experience developing or deploying cloud-native or containerized applications using Docker, AWS, Azure, Kubernetes, or comparable technologies.
Ability to translate technical requirements into production-quality AI software components, integration designs, test cases, and engineering documentation.
Strong analytical, troubleshooting, and communication skills, including ability to collaborate with technical stakeholders to resolve complex integration challenges.
Secret clearance required, with eligibility for TS/SCI.

Nice To Haves

Experience integrating AI applications with Palantir, Databricks, Elasticsearch, or comparable governed enterprise data platforms.
Experience implementing Responder–Evaluator–Reporter architectures, multi-model ensembles, model comparison, or structured AI output verification.
Experience with LLMOps platforms, prompt/model version management, evaluation tooling, and reproducible AI release processes.
Experience supporting Federal Government, Department of Defense, Intelligence Community, Department of Homeland Security, or similarly regulated environments.
Familiarity with Government cybersecurity, privacy, accessibility, data governance, and software authorization requirements.
Experience adapting existing AI algorithms, research prototypes, or software components into operational enterprise applications.
Experience supporting technical demonstrations, Government acceptance testing, operational pilots, or production AI system sustainment.

Responsibilities

Design, develop, integrate, and evaluate advanced AI and LLM capabilities supporting DHS.
Develop and integrate multi-agent AI workflows, RAG, intelligent document-processing capabilities, and AI evaluation frameworks.
Build AI capabilities for investigative analysis, information retrieval, report generation, and human decision support.
Develop, integrate, or evaluate enterprise LLM-powered and generative AI applications.
Implement AI applications using multi-agent orchestration, grounded AI and RAG, or AI assurance, evaluation, and observability techniques.
Integrate AI applications with enterprise data platforms, APIs, or retrieval services, including secure access to structured or unstructured data.
Design and implement secure AI applications with authentication, authorization, permission-aware data access, sensitive-data handling, and human oversight.
Evaluate AI application reliability, including response quality, factual grounding, hallucination mitigation, error handling, and model performance.
Translate technical requirements into production-quality AI software components, integration designs, test cases, and engineering documentation.
Collaborate with AI architects, data engineers, Government stakeholders, and other technical stakeholders to resolve integration challenges.