Expense & travel assistant: travel cost -23%
Trip request, booking, expense and reconciliation handed to an AI assistant — zero receipt pasting for employees.
Expense & travel
Internet · 蓝湖网络 · 2024
Blue Lake Internet has 5,000+ employees, averaging 12K business trips per month. The integration of 4 OA systems, 3 travel platforms and 2 payment rails is reconciled manually, with average reimbursement cycles of 14 days.
Month-end close takes 15 days, overspend happens too often, and employees complain that pasting receipts is worse than writing code. We needed an AI assistant that fully automates request → booking → expense → reconciliation.
How Ouryun ships Expense & travel for 蓝湖网络
Agent workflow as the spine; AI Gateway connecting 4 OAs, 3 travel platforms and 2 payment rails; governance as the budget gate, rule engine, anomaly detector and auto-approver; apps for employees, finance and executives.
Agent workflow
Trip request → best-price booking → auto expense → finance reconciliation as one automated Agent — zero friction for employees.
AI Gateway unified channels
DingTalk, WeCom, Lark, Ctrip, Fliggy, 12306, Didi, Amap and WeChat Pay all flow through one gateway, routed by tenant.
Budget gate + rule engine
Overspend auto-blocked, special approvals one-click, rules hot-updated with no redeploy, anomaly orders auto-escalated.
Model platform
A primary LLM for semantic understanding, plus a specialized small model for receipt recognition and compliance — sensitive data masked.
Three surfaces
Employee-side AI assistant, finance-side auto-audit, executive-side live dashboard — budget, anomalies and volume at a glance.
From connect to ship, in four steps
Connect
In 2 weeks we adapt 4 OAs, 3 travel platforms and 2 payment rails through API / MCP, and inventory every governable expense rule.
Govern
In 3 weeks we ship the budget gate, expense rule engine, auto-approval flow, and sensitive-field masking / anomaly detection.
Deploy
In 3 weeks the Lark / WeCom entry is live; we pilot in 3 departments first, syncing employee + finance feedback biweekly.
Scale
In 8 weeks we roll out to the whole company, retune monthly, and compress month-end close from 15 days to 2.
AI handles trip request, booking, expense and reconciliation — month-end close from 15d to 2d.Full case study delivered and reviewed by the Ouryun team.
Numbers that prove value
Bring this capability into your business
More Ouryun case studies
Manufacturing AI quality inspection
An AOI + AI inspection appliance for PCB production that performs AI-based second-pass review on AOI NG images and reduces manual review load.
Domestic AI inference resource pool
A unified domestic AI inference resource pool for a large manufacturing group, supporting multiple large models and group-level AI platform operations.