An AI-powered conversational interface designed
Aurora is an AI-powered conversational interface designed to automate brand communication while maintaining a human tone. The project explores how generative AI can improve customer engagement and support through scalable interaction systems.
The project focused on designing a scalable chatbot interface capable of automating marketing and support workflows while maintaining a human-centered tone. Through market analysis, interaction modeling, and usability testing, the system was structured to optimize operational efficiency and reduce friction in repetitive processes.
The final outcome is a conversational interface prototype combining AI automation, strategic microcopy, and a scalable design structure ready for multi-channel integration (Instagram, WhatsApp Business API, and email automation).
+35%
+40%
+45%
Problem
Brands struggle to maintain authentic communication with their audiences while scaling digital interactions across social platforms. Manual responses and fragmented tools make it difficult to maintain consistent engagement and timely support.
Users
Primary users:
Social media managers
Marketing teams
Community managers
Goals
automate repetitive interactions
maintain brand voice
increase engagement with followers
Research
I conducted a market analysis of conversational interfaces and AI chatbots used in digital marketing.
Technical Execution
To explore visual identity, I trained a LoRA model using a dataset of images and configured parameters in Google Colab to generate a consistent AI persona. This allowed Aurora to function as a digital brand representative capable of interacting with audiences.
What I learned
This project reinforced the importance of simplicity in UX design.
By focusing on essential interactions and removing unnecessary elements, the experience becomes clearer and more engaging for users.