How I built the Governance Ledger demo
The other three demos on this site take an input and run it through a model live. This one doesn't take an input at all — it's a real, dated excerpt from the routing ledger that Tioga's own AI infrastructure (JARVIS) writes to on every call it makes, anywhere, for any purpose. The interesting engineering here isn't a prompt; it's what the ledger had to look like for this page to be honest.
Last reviewed
A snapshot, not a ticker — on purpose
The 16-row ledger and the "live gateway snapshot" stats above it are hardcoded, dated, and captured at two different times (the ledger rows span Sep 8–9, the snapshot is Sep 9) rather than fetched from a live endpoint on page load. That's a deliberate tradeoff, not a shortcut: a public demo page that live-queries an internal cost/routing gateway is an unnecessary exposed surface for zero real benefit — visitors don't need millisecond freshness on someone else's infrastructure spend, they need to see the shape of what gets logged. The two timestamps are labeled separately in the UI instead of merged into one implied "live" number, so the distinction between "real excerpt" and "real-time" stays honest rather than blurred for effect.
The NIST mapping falls out of the schema, not the copy
GOVERN / MAP / MEASURE / MANAGE aren't a label applied to this page after the fact — they're fields the gateway already records on every call: policy: budget.json for GOVERN, model → served_model for MAP, token/cost/quality fields for MEASURE, and a reserve-then-charge budget check for MANAGE. The page just renders what the ledger schema already tracked. That ordering matters for the offers this demo backs (governed ERP write-path, insurance-underwriting evidence, cost/model governance assessments) — the pitch is that governance evidence is a byproduct of how the infrastructure is built, not a report generated to satisfy an auditor after the fact.
What "requested → served" is actually showing
This window looks different from the one this page originally shipped with (captured Jul 2026), and that difference is the point, not an embarrassment to hide. Back then, most calls resolved to a free local/Google backend before touching billed credit. By Sep 8–9, real call volume has grown past what those free backends alone can absorb, so 15 of the 16 rows in this window route to paid OpenRouter backends (glm-5.2, gpt-terra) — only 0 of 16 (0%) still settle at exactly $0 in this particular window. The mechanism hasn't changed: every request still records what was requested and what actually served it, and the routing policy still tries free/cheap backends first — it's just that there's more real traffic now than the free tier alone covers. Total spend for this 16-call window is still a rounding error against the $30 cap (see the strip above), which is the actual claim this page makes — not that free tier absorbs everything forever.
Why this is the one demo with no prompt-injection surface
Every other demo on this site accepts either a constrained form or a file upload, and the corresponding writeup spends real space on how untrusted input is validated before it reaches a model. This page has no input field and calls no model at request time — it renders a static array shipped in the page bundle. There's nothing to sanitize because there's no path from a visitor's browser to a prompt at all. That's not a gap in this demo; it's the correct shape for what it's actually demonstrating — operational governance data, not a live inference endpoint.
Labeled benchmark
Run 2026-09-09Model
No model call — static ledger excerpt (governance/observability data only)
Sample size
16 logged calls, unsampled (every call in the captured window, not a spot check)
Data source
Real excerpt from Tioga AI's own JARVIS routing gateway ledger — not synthetic, not a demo dataset.
Calls logged
16 (unsampled)
Free-tier resolution
0/16 calls (0%) settled at exactly $0 via local/free-tier routing this window
Ledger window
Sep 8–9, 2026
Snapshot captured
Sep 9, 2026
Limitations
- —This is a fixed, dated snapshot, not a live feed — updated manually when the page is refreshed, not real-time.
- —16 calls is Tioga's own internal AI-operations volume in this window, not a claim about the scale a client engagement would produce.
- —Free-tier resolution varies by window with real call volume — an earlier Jul 2026 capture of this same ledger settled 12/17 (71%) at $0; this window settled fewer because volume grew past what the free backends absorb, not because the routing policy changed.
- —Demonstrates the pattern (governance data as a byproduct of routing infrastructure), not a benchmark of model accuracy or task performance — there's no task being scored here.