Overview
Peraton is seeking a Senior AI Engineer to to design and build production-grade AI systems and lead the next evolution of software delivery across Defense & Health programs by operationalizing AI at scale. This role is focused on embedding AI across the Software Development Life Cycle (SDLC) focused on LLM integration, agent-based systems, and AI-native software engineering, DevSecOps with AI —transforming how systems are built, tested, secured, and operated via AI driven development.
You will design and implement AI-orchestrated, agent-driven workflows leveraging cloud-native platforms and secure government AI environments (including GenAI.mil). The objective is to move beyond isolated AI use cases and deliver repeatable, governed, and measurable AI-enabled systems that accelerate delivery of to scalable, mission-ready AI solutions.This is a engineer role for someone who understands that real impact comes from orchestrating models, data, and workflows into production-grade capabilities.
This position will report to Reston, VA with occastional telework options.
What You’ll Do
- Architect and implement AI-enabled DevSecOps pipelines that accelerate code generation, testing, security, documentation, and deployment
- Design and build LLM-powered applications and agentic systems for software development, testing, security, and operations
- Design and operationalize agentic, multi-step workflows (e.g., code → test → validate → deploy) with appropriate human-in-the-loop controls
- Leverage and integrate GenAI.mil models and commercial LLMs with cloud-native AI services into secure, scalable development environments
- Build and integrate AI microservices and APIs into cloud-native platforms
- Build future-state architecture and data pipelines that ground AI outputs in authoritative, mission-relevant data
- Establish prompt frameworks, chaining strategies and reusable AI patterns that scale across teams and programs
- Integrate AI into IT operations (ticket triage, root cause analysis, observability, incident response) to enable closed-loop automation
- Define and track performance metrics (cycle time, defect reduction, cost-per-feature, SLA improvements) tied to AI adoption
- Lead technical adoption across teams, mentoring engineers and standardizing best practices
- Ensure compliance with federal security, data governance, and AI usage policies
- Implement RAG architectures using mission data (codebases, documentation, operational data) to ground AI outputs
Critical Skills: AI Orchestration & Systems Thinking
LLM & Agentic Workflow Development
- Design and implement multi-agent orchestration, tool integration and workflow automation with tool use, memory, and feedback loops
- Balance automation, control, and reliability in mission-critical environments
- Prompt engineering, prompt chaining, and reusable prompt architectures
- Evaluation frameworks for output quality, reliability, and drift
Data & Retrieval Strategy
- Build and optimize RAG architectures and secure data access patterns
- Structure and govern data (codebases, runbooks, tickets, documentation) for effective AI consumption
- Design, build and maintain Vector databases and semantic search
- Ensure data lineage, integrity, secure access patterns and classification compliance
Model & Platform Orchestration
- Orchestrate across multiple models and endpoints, including GenAI.mil
- Implement routing, fallback, and optimization strategies based on latency, cost, and accuracy
- Design for secure, compliant AI usage in federal environments
Prompt Systems & Evaluation
- Develop scalable prompt frameworks (templates, chaining, reuse)
- Implement evaluation pipelines to measure output quality, drift, and reliability
- Ensure outputs are traceable, testable, and auditable
AI-Enabled DevSecOps, SDLC & AIOps
- Embed AI into CI/CD, security scanning, testing, and documentation workflows
- Apply AI to operations (incident response, anomaly detection, automated remediation)
- Enable closed-loop systems (detect → decide → act)
- AI-assisted SDLCdevelopment workflows and pipeline integration (code, test, security, documentation)
Observability, Metrics & Governance
- Define KPIs tied to AI-driven performance gains
- Implement monitoring for AI system behavior, cost, and outcomes
- Align with DoD/DHA governance, security, and compliance frameworks
What Success Looks Like
- 20–40% improvements in in SDLC cycle time through AI-enabled workflows
- Deliver production-grade AI applications and agentic workflows deployed in secure environments
- Improve code quality, operational efficiency, and system resilience using AI
- Standardized, reusable AI orchestration patterns deployed across programs
- Measurable improvements in SLA performance, cost efficiency, and mission delivery speed
