Building Ride, a robust AI shopping assistant for a leading micromobility review site, Ride Review, has presented exciting opportunities and challenges. The journey highlights the complexities of coding, practical application of AI, and the future potential of personalized shopping experiences.
Launched in 2023, Ride is a pioneering AI shopping assistant developed by BrXnd for the prominent micromobility review platform, Ride Review. It navigates the vast array of product reviews to guide customers towards their ideal purchase. The development journey of Ride, which saw the bridging of theory and practical AI application, underscores the transformative potential of AI in reshaping customer experiences. The project marks a significant step towards the fusion of AI and marketing, heralding a new era of personalized, AI-assisted shopping.
In an industry driven by continuous innovation and experimentation, BrXnd has risen as a leader, exploring the confluence of brands and AI. The brainchild of Noah Brier, BrXnd's central ethos, is to delve into the realities of using and building AI tools, focusing on practical applications over mere theory.
One of BrXnd's explorations has been developing "Ride," an AI shopping assistant for Ride Review, a prominent micromobility vehicle review platform. Launched in June 2023, Ride aims to simplify the customer journey, helping potential buyers navigate thousands of reviews for e-bikes, scooters, and related products.
The initial vision for Ride seemed straightforward: build an assistant that could guide customers to make the right choice for their micromobility needs. However, translating this concept into code brought its share of complexities. A key challenge was to create an AI system capable of understanding the context of the vast data from the review site.
Initially, Noah planned to utilize vector databases to find the nearest matches based on generated embeddings. However, it became clear that more than similarity was needed for an AI shopping assistant. Buyers needed more specific search criteria—if they want an e-bike, they don’t want to be recommended a scooter—thus, the AI had to incorporate effective filters to provide suitable matches for vehicle requirements.
Another hurdle was optimizing the performance of the AI system while ensuring a smooth user experience. The AI was required to perform dual tasks: respond to the user and interpret their response into a preferences taxonomy.
Noah found a solution to these challenges through continual testing, adjustments, and learning. He configured the AI to quickly represent user preferences, which are used to search the database and return results. These results are then incorporated into the bot's response, resulting in a seamless and efficient user experience.
Ride epitomizes the potential of AI to simulate an experienced salesperson's guidance. Notably, the journey of developing Ride underscored the significant role of code in making AI work efficiently in real-world applications. Translating the data returned from the AI into usable results required significant coding work, highlighting the balance between theory and practice when it comes to AI.
Following its launch, Ride has begun to transform the shopping experience on the Ride Review platform. Its development has been an enlightening journey, bringing to the forefront the complexities, challenges, and possibilities of merging AI with practical applications.
Sign up for our mailing list to keep up with what we're up to and be the first to know about new Marketing X AI events.