Infrastructure Architect
Dealstitch AI
See how well this job matches your profile
Sign up to get an AI match score and generate a tailored application in seconds.
Get your match scoreTags
About the role
Role Overview
You will own the cloud, identity, and security layer that enables AI agents to run safely in production. Agents connect to a firm’s systems via MCP servers and must act with the same permissions as the user who invoked them—without creating new security exposure.
Responsibilities
- Own deployment and hosting of agents and MCP servers across Azure, AWS, and Google Cloud.
- Implement identity and access so agent actions run under the invoking user’s permissions, using SSO/OAuth and Microsoft Entra (avoid broad service accounts).
- Build secure connectivity and isolation for private systems, including:
- Private endpoints
- Sandboxes
- MCP tunnels (tunneled connectivity)
- Manage secrets/credentials and enforce least-privilege access.
- Build logging and observability to provide a reliable audit trail of agent activity, while avoiding persisting PII outside the client environment.
- Partner with client IT and security teams to meet requirements such as M365/identity setup, data residency, and security review.
Requirements
- Strong cloud/platform engineering experience across at least two major clouds (Azure, AWS, or Google Cloud), with deep expertise in one.
- Hands-on identity and access experience with:
- SSO
- SAML
- OAuth
- Microsoft Entra / M365 administration
- Security-first mindset: least privilege, secrets management, network isolation, audit logging.
- Ability to own deployments end-to-end and navigate client IT/security review processes.
- Familiarity with:
- Containers (Docker, Kubernetes)
- Infrastructure as Code (Terraform or similar)
- CI/CD
Preferred
- Experience in financial services or other regulated environments.
About Dealstitch AI
Dealstitch AI is a technology company focused on running AI agents in real production environments for private-market organizations. The role centers on building secure infrastructure that connects agent execution to client systems while maintaining identity, permissions, and auditability requirements. The company operates in the intersection of AI engineering, cloud infrastructure, and enterprise security.
Scraped 7/22/2026