Swiggy Introduces AI-Powered Neural Search for Personalized Recommendations and Voice Typing

• ByStartupStory     |    July 21, 2023

Food and grocery delivery giant Swiggy is embracing deep tech with its latest AI-powered neural search feature on its app. The new feature, set to launch in pilot mode, will revolutionize the ordering experience by enabling voice typing and tailored recommendations for food and grocery deliveries. Swiggy’s Chief Technology Officer, Madhusudhan Rao, aims to make the app’s search process fun, effortless, and relevant for users.

In September, users will enjoy the freedom to use open-ended and conversational commands, such as “show me vegan dishes available nearby” or “show me healthy beverage options,” to receive customized and relevant search results.

With the help of a Large Language Model (LLM) adapted to understand dish, recipe, and restaurant terminology, along with Swiggy-specific search data, the neural search tool offers real-time and efficient exploration of food options. It was also trained on the vast food catalog, encompassing 50 million items, to ensure accurate and timely responses to food-related queries.

Swiggy aims to expand the generative AI-backed feature to all its verticals, including food delivery, Instamart, and dining out. They plan to introduce a conversational bot acting as a virtual concierge to guide users through their preferences, moods, ratings, location, and pricing.

Beyond enhancing the user experience, Swiggy is exploring the implementation of generative AI co-pilots for different stakeholders, including delivery partners, restaurant partners, fleet managers, and customer service agents, to improve workflow efficiency.

While other foodtech companies have introduced AI-powered features, Swiggy differentiates itself by creating a tool that truly understands the complexities of food taxonomy. The models go beyond tailored recommendations to offer additional information like ingredient lists and restaurant suggestions.

Additionally, Swiggy is looking to enhance customer service through a GPT-4 powered chatbot developed in collaboration with a third party, aiming to provide efficient and empathetic responses to common customer queries. The company sees its generative AI journey as just beginning, envisioning a future where these methods synergize with human intelligence to serve customers better and unlock significant business value.

Follow Startup Story