AI/ML Platform Engineer

SAIC is seeking an AI/ML Platform Engineer to build and operate governed platforms that enable teams to develop traditional ML, Generative AI, and agentic AI solutions. The role includes platform engineering, customer engagement from concept through production, and integrating emerging AI/ML capabilities for civilian government agency programs.

Responsibilities:

• Enable AI/ML platform services as a governed, multi-tenant enterprise offering tenant onboarding, entitlements, access controls, compute environments, and service-level guardrails spanning traditional ML, GenAI, and agentic AI
• Enable GenAI application building blocks governed access to frontier models via an AI gateway layer, RAG infrastructure (knowledge bases, vector databases, embedding pipelines), per-tenant guardrails, grounding validation, and feedback mechanisms so customers can build GenAI applications on a trusted foundation without reinventing the plumbing
• Enable agentic AI patterns orchestration frameworks, tool/API integration, memory and state management, human-in-the-loop controls, and safety guardrails  so customers can build and run agent-based workflows within the platform's governance boundary
• Enable traditional ML operationalization governed training environments, model deployment pipelines with TAI controls (bias, fairness, explainability, model cards), and fleet-wide drift and quality monitoring across all operationalized models
• Partner with customers from concept to delivery own the engagement end to end: requirements, specs, solution design, hands-on implementation, and production handoff. Sit with mission teams, understand their problems, and help them build on the platform
• Rapidly evaluate and integrate new platform capabilities stay current with the evolving AI/ML landscape, assess new tools and services, and bring the right ones into the platform offering at pace
• Build platform onboarding patterns and IaC templates (Terraform, CloudFormation) with SSO integration, designed for repeatability and portability across deployments
• Develop secure API-based model access and inference endpoints consistent with governance and security standards
• Document reusable patterns, frameworks, and APIs everything built is a starting point for the next deployment, not a one-off