AI Solution Architect

We are seeking an AI Solution Architect to design and lead the build of a governed AI platform and agentic delivery factory that teams can run independently after rollout. You will define the reference architecture, security model, and operating standards while guiding a small build team. Responsibilities Define the reference architecture for a governed AI platform and agentic delivery factory Design the platform components, including agent runtime, gateways, plugins, marketplace, secrets, telemetry, observability, and cost controls Document build-versus-managed decisions and secure sign-off at key delivery gates Own the security and identity model, including SSO integrations, API token handling, and least-privilege access for agents Design guardrails for agent actions across source code, pipelines, and healthcare data workflows Design the agentic PDLC flow with human gates for specification, design, merge, and release Connect orchestration to Azure DevOps using work item triggers, CI hooks, and gate approvals Define the context layer standards agents use, including repo mapping, conventions, decision records, and testing practices Implement evaluation strategy in CI with thresholds that promote agents from supervised to shadow to autonomous Set rules for publishing agents to a marketplace, including ownership, versioning, and review requirements Establish token budgets and provide per-agent and per-work-item cost visibility through telemetry Lead two AI-native engineers through design reviews, PR reviews, and targeted hands-on coding Co-build infrastructure in the client AWS account alongside the platform engineer Harden the platform, author a runbook, and transition operations to the client team by the end of Month 4 Present architecture trade-offs and progress in weekly working sessions and gate reviews with senior stakeholders Requirements 3+ years of experience in AI solution engineering and architecture roles Hands-on architecture leadership skills to guide engineers and drive design decisions Project delivery experience in time-boxed engagements with measurable outcomes Strong cloud platform knowledge, including AWS account-level design constraints Strong security and identity knowledge, including SSO, tokens, and least-privilege access Strong DevOps integration skills with Azure DevOps repos, boards, and CI/CD pipelines Upper-Intermediate English proficiency (B2) Nice to have AWS Bedrock experience for model access and governance