Architect (Multi-skill) — AI Transformation Core Team CPG

Confisa International Group•Cincinnati, OH

10d•Remote

About The Position

The Architect (Multi-skill) will own end-to-end technical architecture for the Foundational AI Platform, including agentic orchestration, MCP server design and MCP gateway integration, observability and SRE, ITSM/AIOps, and AI solutions for applications hosted on Azure and GCP. The role will drive joint architecture decisions with the Knowledge Fabric team, client AI platform teams, and the partner ecosystem. The position is part of an AI-led growth initiative focused on expanding the company’s presence across major Communications, Media & Entertainment clients in North America. The company is a global technology consulting and digital solutions provider with approximately 84,000 professionals, more than 740 active enterprise clients, and approximately USD 4.5 billion in annual revenue. The architect will define platform architecture and reusable starter kits, collaborate with the Knowledge Fabric team on retrieval and service contracts, and design agentic and RAG solutions using Azure and GCP services. The role also provides architecture leadership through standards, reviews, mentoring, roadmap collaboration, and support for AI-native software development practices.

Requirements

10 or more years of experience in architecture.
At least 3 years of experience in GenAI or agentic systems, including at least 2 years of live production experience.
Hands-on experience implementing MCP servers and integrating MCP gateways.
Experience with agent frameworks such as LangGraph, AutoGen, Semantic Kernel, or equivalent.
Experience with RAG/GraphRAG, evaluations, and guardrails.
Experience with Azure and GCP AI/data stacks, Kubernetes, and infrastructure as code.
Strong API, security, and observability fundamentals, including OAuth 2.1, OIDC, and mTLS.

Nice To Haves

AIOps or ITSM integration experience, including ServiceNow and observability platforms.
Experience with spec-driven development approaches such as BMAD, Spec Kit, or OpenSpec, and AI coding assistants.
Experience with multi-vendor delivery in regulated enterprises.

Responsibilities

Own architecture of the agent orchestration layer, including multi-agent topologies, planning and reasoning patterns, agent-to-agent communication, human-in-the-loop gates, and fallback and recovery.
Define the Cockpit/control plane, including agent registry, version control, deployment, runtime configuration, policy enforcement, and traceability dashboards.
Define Guardrails and Responsible AI architecture, including PII redaction, content safety, jailbreak resistance, output validation, policy-as-code, and model risk management.
Design agent observability and evaluation architecture, including tracing, token and cost telemetry, golden-set regression suites, drift detection, and A/B harnesses.
Architect reusable Agent and Copilot Starter Kits for AIOps, engineering productivity, and domain-specific copilots.
Create an MCP server reference design covering Resources, Tools, Prompts, and Sampling, and develop an implementation playbook for service-line teams.
Design the MCP gateway for discovery, registry, routing, transport bridging, and multi-tenancy.
Define retrieval, schema, and SLA contracts with the Knowledge Fabric team; integrate GraphRAG outputs into agents and provide feedback on quality signals.
Architect agentic and RAG solutions using Azure and GCP services native to the applications being served.
Address cross-cloud connectivity, hybrid retrieval, model and inference routing, and secrets and identity propagation across estates.
Conduct architecture reviews, establish design standards, and mentor pods across business units.
Co-own the roadmap with the joint PMO and drive adoption of AI-native SDLC practices, including spec-driven frameworks and AI coding assistants.