Senior AI Platform Architect

We are building an AI platform for a large financial services enterprise. It lets developers inside that organization build and deploy agentic AI workflows: Python applications that call LLMs, use tools, and run as agents on Kubernetes. The platform includes an LLM gateway, an agent runtime, MCP servers for tools and data, a skills registry, retrieval, evaluation, observability, and the identity and governance controls a bank needs before it says yes., We are hiring a Senior AI Platform Architect to own the architecture of this platform. You have 10-12 years of experience and you have built platforms before: a cloud platform, a developer platform, a container platform, or an internal platform other engineers build on. You know AWS well and Kubernetes very well, because the platform runs on Kubernetes and so do the agents. On top of that, you understand the concepts that make up an AI platform (MCP, RAG, agent runtimes such as Bedrock AgentCore, skills and tool registries, gateways, evals) and you can design how they fit together.

This is a hands-on role. You will write design documents, review code, build reference implementations, and make the architecture calls that the engineering team follows. You will also present and defend the architecture to the customer’s architects and security teams.

What You’ll Do

• Own the end-to-end architecture of the AI platform: gateway, agent runtime, MCP layer, skills and asset registry, retrieval, evals, observability, identity, and governance.
• Design how agents run on Kubernetes: packaging, scheduling, scaling, isolation, networking, state and memory, versioning, promotion, and rollback.
• Design the AWS foundation the platform sits on: EKS, IAM, networking, private endpoints, and infrastructure-as-code, built for a regulated enterprise with restricted egress.
• Define the identity model for agents: how an agent authenticates as itself, acts on behalf of a user without gaining more access than that user has, and leaves an audit trail.
• Define the developer experience: how a developer in the customer’s organization goes from an idea to a deployed agent, with the right guardrails and the least friction.
• Set the patterns for MCP servers, skills, tools, and prompts so they are reusable, versioned, and governed.
• Write architecture documents and reference implementations, and review the team’s designs and code against them.
• Present the architecture to the customer’s chief architect, platform teams, and security and model risk reviewers, and incorporate their constraints into the design.
• Evaluate new AI platform capabilities (model providers, agent frameworks, runtimes, registries) and decide what belongs in the platform.
• Mentor senior engineers and raise the team’s platform engineering practices.

Artificial Intelligence (AI), Financial Services, Product Demonstration, Cloud Computing, Artificial Intelligence (AI) Agents, Artificial Intelligence (AI), Finance, MCP - Microsoft Certified Professional, Python Programming/Scripting Language, Software Engineering, * 10-12 years of software engineering, with a substantial part of it spent building platforms: cloud, developer, container, or internal platforms, or as a software architect with deep platform knowledge.

• Very strong Kubernetes: you have designed and run production workloads on it, and you understand networking, scheduling, security, multi-tenancy, and operators, not just deployments.
• Very good AWS: EKS, IAM, VPC networking, private endpoints, Terraform or CDK. You have built and operated a cloud platform before.
• Solid understanding of AI platform concepts: LLM gateways, agent runtimes (Bedrock AgentCore or comparable), MCP, RAG and retrieval, skills and tool registries, evals, and AI observability. You can explain how each piece fits and where the risks are.
• Strong distributed systems, API design, and security fundamentals.
• Comfortable in Python, enough to build reference implementations and review code.
• Experience working inside large enterprises, preferably regulated ones (financial services, insurance, healthcare). You know the difference between a policy you can configure around and one that forces a redesign.
• You can write a clear design document and defend a trade-off in front of a client’s chief architect.

Moring.ai builds enterprise AI that runs in production for regulated industries: financial services and other compliance-heavy organizations. We ship AI applications and agents inside our customers’ own cloud environments, wired into the systems they already run. Not slideware, not demos.