The use case: how Staffbase deploys TrueFoundry
Staffbase, an AI-native employee experience platform, uses the TrueFoundry AI Gateway as a single endpoint for all its AI inference. The gateway pools model deployments across multiple Azure subscriptions, regions and AWS Bedrock, balances load around rate limits, routes traffic by region for data residency, and gives per-feature usage and cost visibility.
More reported results
Figures published in TrueFoundry’s case study. They are not independently measured by usedby.
Since going live in February, the Gateway has governed more than 141 million requests — now around 97.8M in a typical month — and processed 29.2 billion tokens (27.3B input and 1.9B output).
Over a 28-day rolling window, availability sits at up to 99.99998% — comfortably past five nines — across every production environment
embeddings account for over 97% of all Gateway traffic
Around 22 distinct product features and teams route their traffic through the Gateway today
In the customer’s words
The Gateway became the layer every AI feature is built on. It gave us one place to run inference at scale, across every region and provider, with the reliability our customers expect — and it did it without our teams having to become infrastructure experts.Other companies using TrueFoundry
Customers across industries, with public evidence.
Adopt AIConfirmed
Aviso AIConfirmed
Aviva CreditoConfirmed
EchoStarConfirmed
ExamRoom.AIConfirmed
FloQastConfirmedGo straight to the source
Based on TrueFoundry’s published case study. We check the public evidence; the results are reported by the vendor.
truefoundry.com- Last checked
- Nov 7, 2026
Open source capture What the story claims, and what we checked
What we compared with the page.
- The figure: 141 millionChecked
Printed word for word on the page, near the name of Staffbase.
- The passage quoted aboveChecked
Copied word for word from the page, near the name of Staffbase.
- Staffbase uses TrueFoundryChecked
Confirmed line.
- The result itselfNot checked
We quote it; we did not measure it.
Since going live in February, the Gateway has governed more than 141 million requests — now around 97.8M in a typical month — and processed 29.2 billion tokens (27.3B input and 1.9B output).
