Customer support · Production

RailYatri modernizes travel platform with Google Cloud, achieving 60% faster provisioning and AI-powered customer service

The problem

After the pandemic, RailYatri experienced 15-20% monthly growth that its existing cloud provider could not handle, resulting in unplanned downtime and disproportionate database costs, while its legacy on-prem infrastructure was also too rigid to scale with demand.

First attempt

RailYatri's previous cloud provider was unable to scale with rapid post-pandemic growth, causing unplanned downtime and inflated database costs.

Workflow diagram · grounded in source
1
Compute Engine infrastructure migration
integration
“RailYatri now hosts its services, including MySQL and MongoDB, in Compute Engine VMs, establishing a stable and scalable backend that ensures 24/7 availability for ticket booking and other operations”
2
Real-time booking data sync
integration
“when a customer makes a booking, the data is instantly synced across the platform. This real-time capability allows a bus captain to access an updated manifest on their IntrCity Crew App, enabling them to check tickets and verify passeng…”
3
Advanced Resource Period alerts
output
“allows customers to subscribe to alerts for peak travel periods and receive immediate notifications as soon as bookings become available”
4
BigQuery analytics and reporting
integration
“The team uses BigQuery for comprehensive data analysis and shares these insights with business and product teams through Looker Studio”
5
Speech-to-Text call transcription
ai_action
“RailYatri uses Speech-to-Text AI to transcribe customer service calls. This transcription enables the customer service team to analyze call content for keywords, identify common issues, and gauge customer sentiment”
6
Text-to-Speech bus status audio
ai_action
“RailYatri employs Text-to-Speech AI to convert text-based information about bus statuses into spoken audio. This audio is then delivered to customers over the phone, serving as a crucial customer service application. This allows RailYatr…”
7
Insight-driven product improvement
feedback_loop
“by detecting that many bus passengers struggled to locate their boarding points, IntrCity SmartBus was able to implement a direct solution: including an image of the boarding point and Google Maps links on booking tickets to help travele…”
Reported outcome

After migrating to Google Cloud, RailYatri achieved 60% faster server and container provisioning, a 10% increase in holiday bookings via the Advanced Resource Period feature, and significantly improved customer service through Speech-to-Text and Text-to-Speech AI.

Reported metrics
Server and container provisioning speed60% faster
Holiday bookings increase10%
Monthly platform growth15-20%
Reported stack
Compute EngineBigQueryLooker StudioGoogle Speech APIsSpeech-to-Text AIText-to-Speech AIGoogle MapsMySQLMongoDB
Source
https://cloud.google.com/customers/railyatri
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

After migrating to Google Cloud, RailYatri achieved 60% faster server and container provisioning, a 10% increase in holiday bookings via the Advanced Resource Period feature, and significantly improved customer servic…

What tools did this team use?

Compute Engine, BigQuery, Looker Studio, Google Speech APIs, Speech-to-Text AI, Text-to-Speech AI, Google Maps, MySQL, MongoDB.

What results were reported?

Server and container provisioning speed: 60% faster; Holiday bookings increase: 10%; Monthly platform growth: 15-20% (source-reported, not independently verified).

What failed first in this deployment?

RailYatri's previous cloud provider was unable to scale with rapid post-pandemic growth, causing unplanned downtime and inflated database costs.

How is this customer support AI workflow structured?

Compute Engine infrastructure migration → Real-time booking data sync → Advanced Resource Period alerts → BigQuery analytics and reporting → Speech-to-Text call transcription → Text-to-Speech bus status audio → Insight-driven product improvement.

Related customer support cases