The assistant, an upgrade to MakeMyTrip’s existing AI agent Myra, is designed to help users across the full span of travel planning from destination discovery to booking, and post-sales support. It can respond to both voice and text inputs, allowing users to interact naturally in either language. According to the company, the assistant can handle open-ended and complex travel-related queries such as:
- “Where can I go in August for a relaxing holiday with my kids?”
- “Mujhe Udaipur mein 3-star hotel 3500 ke budget mein chahiye”
- “I want to go to south India to cover Madurai, Rameswaram, Kovalam, Kodaikanal. Can you suggest me the best route? I don’t want to travel via flight.
What sets Myra apart, the company claims, is its ability to not only provide suggestions but also carry the user through to an actual booking within the same conversational flow.
Built on a system of specialised AI agents across key travel services flights, hotels, holiday packages, ground transport, visas, and forex Myra supports multimodal input, including voice, text, and eventually images and videos. It’s also designed to handle follow-up questions, itinerary changes, and support requests within the same thread.
According to Rajesh Magow, MakeMyTrip’s Co-Founder and Group CEO, the launch is aimed at making the travel booking experience more intuitive and inclusive. He notes that this could help bridge the gap for users in non-metro areas who may not be as comfortable with English or structured app interfaces.
Sanjay Mohan, Group CTO, described the system as one of the company’s most complex tech builds to date, involving custom language models layered with planning and verification systems. The system is still learning from real-time interactions, with ongoing development expected.This release builds on MakeMyTrip’s earlier AI efforts, including voice-led support, predictive tools for train travel, and features like Fare Lock and Zero Cancellation. The next phase of development is expected to focus on semantic search and AI use cases involving images and video, to enable more abstract and intent-driven interactions.
