AI Architecture & Governance Leader Enterprise AI Platforms
Architecture & Platform Define end‑to‑end AI solution architectures (cloud & on‑prem) including model serving, RAG/LLM patterns, vector indexing, data integration, and observability. Establish reference architectures, “golden paths,” and reusable templates that integrate with the enterprise AI platform. Lead evaluations and POCs of AI capabilities (LLM serving engines, vector DBs, orchestration frameworks, evaluation toolchains, guardrails). Partner with Enterprise Architecture to align AI patterns with enterprise standards, security, and roadmaps. Guide the design of scalable inference topologies (GPU/CPU, autoscaling, caching, batching, token optimization) and performance tuning. Implement model/data registries, adapter/prompt catalogs, and change control with traceability from use case → model → dataset → deployment. Ensure AI supply‑chain security (licenses, provenance, SBOMs, model signing), privacy, and regulatory compliance. Establish patterns for secure tool invocation, approvals, auditability, and exception handling across business processes. Build cost governance and FinOps practices for AI workloads (token usage, GPU utilization, autoscaling policies). Run intake, triage, and prioritization of AI and agentic automation use cases; align with business OKRs and platform strategy. Shape success metrics and delivery roadmaps in partnership with product, data, security, and engineering teams. Drive build/partner/buy analyses and vendor selections; negotiate guardrail requirements and SLAs. Provide hands‑on guidance to product squads on decomposition, MVP scoping, and path‑to‑production. Evangelize best practices, create enablement materials, and mentor architects/engineers and product managers. Drive alignment across security, data, platform, and enterprise architecture; foster a culture of responsible innovation. 10+ years in software/AI/ML engineering, platform or enterprise architecture, with 5+ years in a leadership role managing cross‑functional initiatives. Engineering degree (Computer Science, Electrical/Computer Engineering, or related). Proven experience defining AI solutions architectures (cloud & on‑prem), including LLM/RAG patterns and model lifecycle. Strong understanding of AI inference—throughput/latency trade‑offs, batching/caching, GPU/CPU sizing, quantization, token optimization. Demonstrated Enterprise AI Governance experience (policies, approvals, model/data lineage, risk/compliance, Responsible AI). Hands‑on with Kubernetes (Helm/Kustomize, autoscaling, service mesh, GPU operators) and LLM serving engines (e.g., vLLM, TensorRT‑LLM, Triton, KServe/Seldon, Ray Serve). Experience with agentic automation frameworks (e.g., LangGraph, Semantic Kernel, AutoGen) and RPA (e.g., Microsoft Power Automate, UiPath, Automation Anywhere). Excellent full‑stack web & mobile architecture knowledge (APIs, eventing, microservices, identity/authorization, mobile backends). Experience as a Technical Product Manager or close TPM partnership—portfolio planning, vendor evaluation, and stakeholder management. Working knowledge of the enterprise IT ecosystem (identity, networking, security, data platforms, DevSecOps, compliance). Strong communication and executive‑level storytelling; ability to influence and drive consensus across diverse stakeholders. Familiarity with Enterprise Architecture frameworks and tools (e.g., TOGAF, Zachman; LeanIX/Ardoq/Sparx EA). Experience operating AI platforms at scale (multi‑tenant, multi‑cloud/on‑prem), including GPU scheduling (NVIDIA GPU Operator/MIG) and edge/hybrid scenarios. Knowledge of MLOps/LLMOps toolchains (MLflow, Databricks/Mosaic AI, Vertex AI, Azure AI/ML, SageMaker; model/data catalogs and evaluators). Experience with vector databases and RAG components (e.g., Azure AI Search, Pinecone, Weaviate, Milvus), and feature stores (e.g., Feast). Observability expertise (OpenTelemetry, Prometheus/Grafana) and AI quality monitoring (e.g., human feedback, eval pipelines, drift detection). Security, privacy, and compliance background (policy‑as‑code with OPA/Kyverno, model/content safety, data masking, DLP, encryption). Certifications: TOGAF, CKA/CKS, major cloud AI certifications (Azure/AWS/GCP), or Responsible AI training. Experience establishing governance councils and federated operating models across business units. Track record delivering agentic automations that integrate with enterprise systems (ERP/CRM/ITSM) with measurable ROI.
