Tanium · AI Gateway Reference Architecture

Secure, policy-governed AI for regulated enterprises.

A working reference lab that shows how identity-aware classification, DLP, model routing, and per-user cost attribution compose into one defensible AI gateway. Walkthrough runs ~5 minutes.

Identity
Keycloak · OIDC · 5 groups

Every request carries a JWT with max_classification, token_budget_daily, and allowed_mcp_servers claims. No claims, no access.

Classification
PUBLIC → RESTRICTED

Prompts auto-classify by content. Tier mismatch returns 403 before any model ever sees the payload.

DLP
Block · Redact · Allow

Request and response sides both scanned. Personal identifiers, secrets, and restricted markers caught before they reach the model — or the user.

Routing
Haiku · Sonnet · Opus

Model selection by prompt complexity and data classification. Cheap for trivial queries, Opus only when reasoning demands it.

Attribution
Per-user cost telemetry

Every request is logged with user_id, model, tokens, and USD cost. Grafana surfaces org-wide and per-user spend in real time.

Tools
MCP · 5 servers

Salesforce, Slack, RAG knowledge, plus mocks for demo flow. Tool access governed by the same identity claims that drive classification.

What the demo walks through

Try the lab yourself → Download video (14 MB)