Role Overview
The Federal Reserve Board offers a Summer 2027 internship for a Super Intelligence (SI) Connectors Platform Engineer. The intern will work in the Application Infrastructure Support unit, building a governed platform connecting SI/AI tools to enterprise data via Model Context Protocol (MCP) servers, RAG services, and identity/gateway layers on Kubernetes (Amazon EKS) in AWS, using infrastructure as code and CI/CD pipelines.
Responsibilities
- Build and deploy MCP servers exposing Board data sources to SI/AI clients using Python and MCP SDK
- Stand up and test RAG components: document ingestion, chunking, embeddings, vector search on PostgreSQL with pgvector
- Deliver platform changes via Terraform, Ansible, GitLab CI, ArgoCD GitOps to Kubernetes
- Contribute to identity/access layer: OAuth client registration, audience-bound tokens, LiteLLM gateway routing
- Maintain MCP connector inventory and catalog
- Write templates, documentation, quick-start guides for enterprise connector standards
- Participate in daily stand-ups, sprint planning, design reviews
Requirements
- Enrolled in undergraduate/graduate degree at accredited university, returning after internship
- U.S. citizenship required
- 0 years experience (entry-level)
- Interest in SI/AI, platform engineering, cloud infrastructure
Learning Objectives
- Build, containerize, deploy working MCP server with authentication and gateway registration
- Implement RAG ingestion/retrieval pipeline on PostgreSQL with pgvector, measure retrieval quality
- Write and run GitLab CI pipeline building container image, automated tests, deploy via ArgoCD
- Author Terraform and Ansible changes delivered through merge requests and code review
- Apply platform security controls (OAuth 2.0 audience-bound tokens, deny-by-default authorization, secrets management, audit logging)
- Produce developer quick-start guide and present completed work