What challenges did Yatra face before AI implementation?

Expense management required high-touch manual reconciliation of receipts across jurisdictions. Standard OCR often failed on multilingual, damaged, crumpled, faded, or handwritten receipts, causing reimbursement bottlenecks and inconsistent fraud detection.

What was the impact of the AI solution?

Receipt scanning latency fell from 20 seconds to 3-4 seconds. The solution automated routine verification and moved Yatra from search to an autonomous assistant. The platform supports over 6,000 corporate customers.

Source

GCPEvidence: Medium50/100

Yatra automates expense logs and builds an agentic travel assistant using Gemini Enterprise + Vision AI (RECAP/DIYA)

Use case typeExpense management automationUpdated Jul 19, 2026

Yatra (India, corporate travel services) built RECAP on Google Cloud to automate manual expense reconciliation and DIYA as an agentic travel assistant for corporate travelers. RECAP uses Vision AI to extract line-item and tax information from complex receipts, while DIYA uses Gemini Enterprise Agent Platform and Gemini to understand travel intent and execute travel-life-cycle tasks. The implementation reduced receipt scanning latency from about 20 seconds to 3-4 seconds and supports more than 6,000 corporate customers.

Organization: Yatra
Industry: Consumer & Food
Location: India
Published: July 2026

Reported outcomes

3-4%

receipt scanning latencyTime & speed

Strategic outcomes

Cost efficiencyAutomated routine receipt verificationCustomer experience & trustShifted to an autonomous travel assistant

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Receipt scanning latency: 3-4% decrease

Google Cloud Customer StoriesJul 19, 2026Customer storyInferred claimMedium evidence strength

reducing receipt scanning latency from 20 seconds to a range of 3 to 4 seconds

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Yatra
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Expense management automation
2Travel planning automation
3Conversational assistants

Expense management required high-touch manual reconciliation of receipts across jurisdictions.
Standard OCR often failed on multilingual, damaged, crumpled, faded, or handwritten receipts, causing reimbursement bottlenecks and inconsistent fraud detection.

Yatra developed RECAP on Google Cloud to automate verification and extract receipt data from complex images.
The company built DIYA as an agentic assistant on Gemini Enterprise Agent Platform using Gemini 3.0 architecture to interpret booking intent and handle complex travel queries.
The system runs on Google Cloud infrastructure including Compute Engine and uses secure data handling for proprietary travel data.

Technologies

Gemini Enterprise Agent Platform Gemini Vision API Compute Engine

Receipt scanning latency fell from 20 seconds to 3-4 seconds.
The solution automated routine verification and moved Yatra from search to an autonomous assistant.
The platform supports over 6,000 corporate customers.

Architecture

Yatra built RECAP on Google Cloud with Vision AI to extract receipt line items and tax data, and DIYA on Gemini Enterprise Agent Platform / Gemini 3.0 to interpret travel intent and handle booking workflows. The solution runs on Google Cloud infrastructure including Compute Engine and uses secure handling for proprietary travel data.

Sources & evidence1

Evidence: Medium50/100Evidence strength