LatestRunning support
AI support for API companies: structure docs for retrieval
How API and developer tool companies should structure docs, error codes, changelogs and status for an AI support agent, and where the limits are.
8 min read
The LayBuild blog
We build an AI agent that answers support questions from your own documents, says so when they don't cover the question, and lets your team take over the chat. These posts explain how that works, where it falls short, and how to measure it honestly.
LatestRunning support
How API and developer tool companies should structure docs, error codes, changelogs and status for an AI support agent, and where the limits are.
8 min read
1 to 9 of 46 articles

Running support
Latency budgets, speech recognition errors, turn-taking and cost structure for voice support agents, and why LayBuild is text-only today.
7 min read

How LayBuild works
Two test layers for an AI support agent: fast contract tests with a stubbed model on every commit, and a golden question set against a real model.
10 min read

Retrieval and grounding
LayBuild replies in the customer's language, but default retrieval is English. What that means for non-English questions and how to make it work.
7 min read

Running support
What to set up before a sale: pinned sale terms, Q&A pairs for shipping and returns, plan limits, rate limits, handoff staffing and a load test.
8 min read

Security and deployment
What an AI agent should and should not do in payments support: no account actions, what output redaction covers, card data and PCI DSS, dispute handoff.
7 min read

Security and deployment
LayBuild isn't HIPAA compliant and signs no BAA. What HIPAA asks of a vendor handling PHI, where we fall short, and safe patterns for healthcare support.
7 min read

Security and deployment
How access control works in LayBuild: Owner, Admin, Agent and Customer, what each role can reach, the audit log, TOTP two-factor, and what's missing.
6 min read

Security and deployment
Run LayBuild with docker-compose: the six services, the env you must set, which calls leave your network, and how to keep model traffic on your own server.
7 min read