CNA
System Architect Director - AI Platform Engineering
Chicago, IL, USA · Director
Sponsorship not specified$97k-$189kDetected 13 days ago
PythonDistributed SystemsCode ReviewBigQueryVector DatabasesGCPCI/CDPlatform EngineeringMachine LearningData ScienceLLMsRAGAI OrchestrationStakeholder ManagementPerformance ManagementLeadershipCommunicationProblem SolvingMentoring
About the role
- You have a clear vision of where your career can go.
- And we have the leadership to help you get there.
- This role combines the strategic depth of a principal architect with the hands-on leadership of a delivery-focused engineering director.
Responsibilities
- Own and continuously evolve the enterprise AI Platform reference architecture, encompassing all critical layers including model serving, orchestration engines, data and knowledge grounding pipelines, observability infrastructure, and ensuring the platform scales reliably to enterprise-grade workloads and usage patterns.
- Define and enforce platform-wide standards, reusable design patterns, and golden-path templates that enable product and feature teams to build, deploy, and operate AI solutions safely, consistently, and with significantly reduced time-to-production.
- Drive end-to-end delivery of new platform capabilities - from initial technical discovery and architecture design through prototyping, hardening, and full production rollout while maintaining meaningful hands-on involvement at critical technical milestones to ensure quality and coherence.
- Build and maintain robust CI/CD and AIOps pipelines specifically designed for AI systems, incorporating automated evaluation gates, model and data versioning controls, staged deployment promotion, and continuous cost and performance optimization guardrails.
- Architect enterprise-grade multi-agent and single-agent workflow patterns for high-value business use cases, establishing clear standards for orchestration design, state and memory management, tool and API integration, and safe autonomy controls including human-in-the-loop approvals, permission scoping, and comprehensive audit trails.
- Design and implement knowledge grounding systems - spanning hybrid retrieval strategies, semantic reranking, ontology-driven entity modeling, and knowledge graph integration - to measurably improve AI output accuracy, traceability, and readiness for regulatory audit.
- Embed responsible AI and compliance-by-design principles into every layer of the platform, covering data privacy protections, enterprise secrets management, granular access controls, output leakage prevention, and model risk governance practices aligned to enterprise and regulatory standards.
- Directly manage, mentor, and grow a high-performing team of platform engineers, solution architects, and technical specialists - hiring hands-on builders, coaching technical leadership skills, and sustaining a healthy innovation pipeline that continuously advances the organization's AI platform maturity.
- Proven agentic system design capability, with hands-on experience architecting multi-agent and single-agent workflow systems using orchestration frameworks such as Lang Graph, Google ADK - including tool and function calling patterns, state and memory persistence strategies, and robust safe autonomy controls.
- Applied GenAI depth spanning LLM solution architecture patterns, model selection and routing strategies, advanced prompt engineering techniques, fine-tuning and RLHF tradeoffs, and production-grade RAG and hybrid retrieval system design and optimization.
Requirements
- Knowledge grounding and semantic layer proficiency, including experience building canonical ontology and entity models, designing vector search and hybrid retrieval pipelines, integrating knowledge graphs, implementing reranking strategies, and establishing citation and traceability mechanisms that support compliance.
- Strong SDLC and hands-on engineering fundamentals, including Python proficiency, architectural and code review practices, comprehensive testing strategies for AI systems, technical debt management, refactoring discipline, and operational readiness standards.
- Strong Python proficiency and deep practical GCP experience - Vertex AI, GCP Agent Builder, and Gemini - with the ability to engage credibly in hands-on technical work alongside the engineering team.
Compensation
- I n certain jurisdictions, CNA is legally required to include a reasonable estimate of the compensation for this role.
Benefits
- Bachelor's degree in Computer Science, Software Engineering, Information Technology, or equivalent required; Master's degree in AI, Machine Learning, Data Science, or related discipline strongly preferred.
- Prior experience in regulated industries (insurance, financial services, or healthcare) strongly preferred, given stringent governance, auditability, and model risk management requirements.
This listing is sourced directly from CNA's careers page and normalized into a canonical job model.