Senior AI Software Engineer

The General Dynamics Information Technology team is seeking a Senior AI Software Engineer to develop, modernize, integrate, and sustain an on-premises, AI-powered enterprise chatbot platform for the U.S. Department of State. The role focuses on building LLM and RAG capabilities, document processing, and containerized inference in highly controlled, classified, air-gapped environments.

Role Responsibilities

• Design, develop, test, and maintain Python-based applications, microservices, and APIs for an on-premises chatbot.
• Extend, configure, and integrate OpenWebUI with enterprise AI services and backend systems.
• Develop custom OpenWebUI Functions, tools, and integrations for mission-specific use cases.
• Integrate approved LLMs using OpenAI-compatible APIs and inference frameworks including vLLM and Ollama.
• Create reusable components for AI model access, prompt management, response processing, and application workflows.
• Collaborate with AI/ML engineers to improve inference performance, response quality, and application reliability.
• Support multimodal AI use cases, including document understanding, optical character recognition, and vision-language model integration.
• Develop backend capabilities supporting secure enterprise AI workflows.
• Design, implement, and optimize RAG workflows for enterprise document search and question answering.
• Integrate document processing technologies including Docling, Tesseract, and vision-language models into automated ingestion pipelines.
• Implement document chunking, metadata extraction, embedding generation, and retrieval workflows.
• Integrate embedding models and reranking models to improve retrieval accuracy and relevance.
• Work with PostgreSQL and pgvector for vector search, semantic retrieval, and document indexing.
• Develop APIs and background processing services for document ingestion and knowledge base management.
• Collaborate with data engineers to support accurate storage, indexing, and retrieval of enterprise information.
• Implement testing and evaluation methods to measure RAG accuracy, retrieval relevance, and answer quality.
• Integrate and maintain AI inference services using vLLM, Ollama, and compatible model-serving frameworks.
• Support deployment and configuration of approved open-weight LLMs, embedding models, rerankers, and vision-language models.
• Develop software integrations using OpenAI-compatible REST APIs.
• Assist in optimizing inference configurations, including context length, GPU memory utilization, concurrency, and model serving performance.
• Support transition from existing inference frameworks to improved architectures as technologies mature.
• Troubleshoot inference errors, model loading issues, and integration failures in coordination with AI/ML engineers.
• Develop benchmarking and performance-testing utilities for AI workloads.
• Support migration of DS Chat components from traditional deployments into containerized services.
• Develop, configure, and maintain container images using Podman or Docker; create and maintain Podman Compose configurations for services, networking, volumes, and environment variables.
• Integrate services including OpenWebUI, vLLM, Docling, PostgreSQL, pgvector, PgBouncer, and Redis.
• Develop scripts and automation for repeatable installation, deployment, upgrades, and recovery.
• Implement health checks, readiness checks, logging, and service dependency management; troubleshoot container networking, authentication, storage, GPU access, and service connectivity.
• Collaborate with infrastructure and cybersecurity teams to improve deployment consistency and security.
• Research, prototype, and develop agentic AI capabilities for future DS Chat releases, including controlled tool-calling and multi-step workflows.
• Develop integrations enabling approved AI models to interact with authorized enterprise APIs and services.
• Explore emerging technologies involving AI orchestration, workflow automation, and Model Context Protocol (MCP).
• Design secure execution boundaries, permissions, audit logging, and approval mechanisms for AI-driven actions, with safeguards supporting human oversight and preventing unauthorized actions.
• Apply software engineering best practices, including modular architecture, code reviews, documentation, and version control.
• Develop automated unit, integration, regression, and performance tests using pytest.
• Build integration tests validating communication among OpenWebUI, inference services, document processing components, and databases.
• Support CI/CD processes appropriate for disconnected and classified environments.
• Troubleshoot production issues using application logs, metrics, traces, and system diagnostics, including operational logging and monitoring with Splunk.
• Support vulnerability remediation, dependency management, software supply chain reviews, and container security assessments.
• Produce technical documentation, deployment procedures, architecture diagrams, and operational runbooks.
• Participate in Agile activities, including sprint planning, technical design reviews, and demonstrations.
• Design software solutions suitable for secure, on-premises, air-gapped government environments.
• Implement secure authentication integration, role-based access controls, API authorization, and data protection measures, ensuring AI services communicate only through authorized internal network paths.
• Support software packaging and offline deployment using preapproved dependencies, container images, and model artifacts.
• Collaborate with cybersecurity personnel to address security findings and support Risk Management Framework (RMF) activities.
• Assist with technical documentation and evidence required for TRB/CCB reviews and Authority to Operate (ATO) processes.
• Follow approved procedures for handling sensitive and classified information.

Required Qualifications

• Minimum Secret Clearance, no sponsorship
• Bachelor's degree in Computer Science, Software Engineering, Artificial Intelligence, Computer Engineering, or a related technical field, or equivalent practical experience
• 5+ years of professional software engineering experience, including backend application development
• Strong proficiency in Python and experience developing production-quality software
• Experience developing RESTful APIs and integrating distributed application services
• Hands-on experience integrating LLMs or generative AI capabilities into software applications
• Understanding of transformer-based models, inference APIs, prompt engineering, and LLM application development
• Experience with RAG concepts, including document ingestion, embeddings, vector search, and retrieval pipelines
• Experience working with PostgreSQL
• Familiarity with container technologies such as Podman or Docker
• Experience with Linux-based development and application deployment
• Proficiency with Git, automated testing, debugging, and software development best practices
• Experience troubleshooting applications across multiple interconnected services
• Strong analytical, problem-solving, communication, and technical documentation skills
• Ability to work effectively with cross-functional teams and government stakeholders

Technology Stack

• Python, OpenWebUI, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG)
• Vector databases, OpenAI-compatible APIs, vLLM, Ollama
• RESTful APIs, OpenAI-compatible REST APIs
• Docling, Tesseract, vision-language models
• PostgreSQL, pgvector
• Podman, Docker, Podman Compose
• Redis, PgBouncer, pytest, Splunk
• Model Context Protocol (MCP)
• Risk Management Framework (RMF)

Position Details

• Location: Sterling, VA (Onsite)
• Salary: USD 128,039 - 173,229 per yearly
• Requisition Type: Pipeline
• Req#: RQ230390
• Public Trust: None
• Work Requirements: 7 + years of related experience (may vary based on technical training, certification(s), or degree)
• Travel: Less than 10% travel required
• U.S. Citizenship: Required
• Clearance Level: Secret

Identity Verification

• Identity verification uses advanced biometrics and artificial intelligence
• Expected to be on camera during virtual interviews
• May take picture to verify identity and prevent fraud
• Authorization required for collection, processing, and use of biometric data for identity verification and security purposes