Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.jingjaiops.com/llms.txt

Use this file to discover all available pages before exploring further.

JingjaiOps embeds Claude AI throughout the workflow — turning plain-English descriptions into structured quotes, summarizing closed jobs, and answering questions about your data without you writing a single SQL query.

Powered by Claude AI

All AI features in JingjaiOps run on Anthropic’s Claude AI. We chose Claude because:
  • Strong Thai-language understanding for bilingual workflows
  • Reasoned, structured output that fits cleanly into business systems
  • Privacy-conscious — your data is processed only for your own queries and is not used to train models
  • Reliable — the API has predictable behavior we can build a product on
Every AI interaction is scoped to your business’s data only. JingjaiOps does not share your inventory, customer data, or pricing across tenants.

Available features

AI Quote Builder

Describe a job in plain language; AI suggests the line items.

AI Job Report

Auto-generated summary of every closed job.

Ask Your Data

Type a question; get a chart, a table, or a number.

Plan availability

AI features are part of the Operations plan and above:
PlanAI Features
StarterNot included
ProNot included
OperationsAll AI features included
EnterpriseAll AI features + custom prompts and bring-your-own-key
If you’re on a Pro or Starter plan and want to try AI features, contact support@jingjaipost.com — we can enable a 30-day trial on your account.

Response time

Most AI calls return in 2–6 seconds. The exceptions:
  • AI Quote Builder — 4–10 seconds (it has to match your inventory)
  • AI Job Report — runs in the background after inspection completes, ready in under 30 seconds
  • Ask Your Data — 3–8 seconds depending on complexity
The UI shows a clear progress state during AI calls. If a call fails (network error, rate limit), it retries once automatically and then shows a friendly error with a retry button.

Privacy and data handling

Your data passes through the Claude API to generate AI responses. Important guarantees:
  • No training on your data. Anthropic does not train future models on customer API data.
  • Per-tenant isolation. Queries from one JingjaiOps account never see another account’s data.
  • Logged for audit. Every AI call is logged in your account’s audit trail (who asked, when, what was asked).
  • Deletable. AI conversation history is wiped when you delete a record (e.g., archive a customer).
For high-sensitivity use cases (e.g., government tenders), the Enterprise plan supports bring-your-own-API-key, where the AI calls go through your own Anthropic account and never touch JingjaiOps’ shared infrastructure.

Disabling AI features

If you don’t want AI features at all, disable them per user in Settings → Users → Edit User → AI Permissions. Disabling on a user means they don’t see AI buttons in the UI and AI features are skipped for their actions. The features can also be disabled account-wide in Settings → Business Settings → AI Features → Off. The data the AI would have generated (job reports, quote suggestions) is left blank.

Cost

AI features are included at no extra cost on the Operations plan within reasonable usage. “Reasonable” means typical business usage — building quotes, generating job reports, asking 20–30 data questions per week. If you use the API extraordinarily heavily (1000+ AI calls per day), we’ll reach out to discuss either an Enterprise upgrade or BYOK. You can monitor your AI usage in Settings → Usage.