Nodo
Year
2026
TYPE
Mobile
CONTEXT
Artificial Intelligence
DURATION
6 months
OVERVIEW
A conversational AI layer for ecommerce, designed as a personalized shopping assistant that enables users to discover, compare and select products through natural language, integrated into mobile apps as a scalable, multi-brand solution aimed at increasing conversion while introducing a more adaptive and context-driven shopping experience.
CHALLENGE
Conversational interfaces introduce a fundamentally different interaction model within ecommerce environments, shifting from structured navigation to open-ended, intent-driven inputs. While this enables more flexible and personalized experiences, it also increases ambiguity, unpredictability, and dependency on data quality.
Nodo was developed within a SaaS ecosystem, where the solution needed to operate across multiple clients, categories, and brand identities. The challenge was to design a system capable of interpreting diverse user inputs and translating them into relevant product suggestions, while maintaining consistency, usability, and visual adaptability at scale.
PROCESS
Conducted usability testing (~30 participants) through interactive prototypes and live environments, evaluating task completion, input behavior, and user expectations when interacting with AI-driven interfaces.
Structured the conversational model around open input, combining AI-generated responses with product recommendations and follow-up prompts to refine user intent.
Designed key interaction states, including empty states, error handling, and fallback scenarios to support system limitations and maintain continuity.
Adapted the interface to operate within a multi-brand environment, ensuring visual neutrality while allowing integration with different design systems.
Collaborated with product and engineering on scope definition, feasibility, and implementation, including the transition from webview to a native mobile architecture.
SOLUTION
Nodo introduces a conversational layer within ecommerce apps, allowing users to search, explore, and select products through natural language interactions. Accessible through a floating entry point, the experience opens into a dedicated interface where users can engage directly with the assistant.
The system combines text responses with dynamic product cards, enabling users to view details, add items to cart, or save products without leaving the conversation. Each interaction is designed to both respond to the initial request and guide the user forward through follow-up prompts, supporting a more iterative decision-making process.
IMPACT
Nodo was launched during VTEX Day 2026 as part of the company’s innovation initiatives, attracting strong interest from clients and serving as a key demonstration of AI applications within the platform.
As an MVP, the release marked the first step in introducing conversational experiences into the SaaS ecosystem, with early adoption driven by client demand and curiosity around AI-assisted commerce. During the launch, the project attracted interest from over 50 clients, including major Brazilian brands across segments such as health and fashion.
The project established a foundation for future iterations, informing both product direction and technical exploration around conversational interfaces, while expanding my experience designing systems that operate across multiple brands, categories, and levels of content maturity.
