Concepts Overview
This section explains how Nexus works under the hood — the architecture, design decisions, and core concepts that power a secure, auditable AI gateway.
Who Should Read This?
Section titled “Who Should Read This?”Contributors, security engineers, operators, and technical decision-makers who need to understand Nexus system design, component interactions, and architectural tradeoffs.
What’s Covered
Section titled “What’s Covered”Nexus is built as Rust microservices on an open-source data plane (Postgres, ClickHouse, S3-compatible storage, NATS, Redis):
- Request flow through gateway, policy engine, and provider routing
- Multi-tenant identity model (workspaces, organizations, projects)
- Tamper-evident audit chains with cryptographic proof
- Data models and persistence layer
- Architectural decisions (ADRs) and their rationale
- Artifact storage for files and vector stores
In This Section
Section titled “In This Section”- Data Models - Postgres and ClickHouse schemas, trace contracts, and persistence architecture
- Tracing and Auditing - How Nexus separates audit accountability, analytics metadata, and opt-in trace payloads
- Architecture Decisions (ADRs) - Key technical decisions including Rust microservices, OpenAI-compatible API, tamper-evident audit, and event messaging
- Artifact Storage - OpenAI-compatible file and vector-store lifecycle with pgvector and object storage