Senior Technical Architect – Secure AI Infrastructure (LLM/MCP Gateways & Agent Sandboxing)

Project overview

For an AI security program at a large enterprise in Germany, we are looking for an experienced, hands-on Technical Architect to design secure and scalable AI infrastructure.

The focus is on LLM gateways, MCP (Model Context Protocol) gateways and isolated execution environments for AI agents across hybrid cloud and on-premises infrastructure.

Requirements

Several years of experience as a Technical Architect, Solutions Architect or Senior Systems Engineer, with a strong focus on security and integration.

Deep knowledge of networking, identity and access management, service meshes, virtualization and Linux.

Hands-on experience with hybrid deployments, Docker, Kubernetes, API gateways, proxy architectures and traffic management.

Practical experience with LLM deployment patterns, API security for AI, MCP and agent sandboxing.

Familiarity with isolation technologies such as Firecracker, gVisor, Kata Containers or comparable microVM and container solutions.

Proven experience writing whitepapers and communicating complex technical concepts to different audiences.

Fluent English.

Interested?

Please apply directly through the platform and include:

Your current CV

Two project references demonstrating comparable AI infrastructure work, preferably involving LLM/MCP gateways or agent sandboxing

Your earliest start date and available capacity

Your day rate and conditions for occasional on-site work in Germany

Project details

As soon as possible

Until approximately September 2027

At least 65 project days; expected utilization approximately 50%

Remote, with occasional on-site meetings in Germany

Project language: English

Engagement: Freelance project assignment

Your responsibilities

Design secure end-to-end architectures for LLM gateways, MCP gateways and isolated AI agent execution.

Develop hybrid cloud and on-premises designs aligned with enterprise security and data privacy requirements.

Create whitepapers, technical guidelines and reference architectures for engineering teams, security specialists and non-technical stakeholders.

Advise on integration scenarios, including legacy systems connecting to AI models, multi-tenant agent execution and zero-trust network boundaries for AI workflows.

Collaborate closely with AI security working groups, enterprise architects and engineering leads.

• Start: ab sofort
• Location: DE
• Rate: 80-80 EUR/h
• Languages: English