Meet Dana,
Our AI Business Agent

Lumi was developed as an early experimental AI assistant for Retune, combining a conversational interface with a 3D customer-facing avatar. The objective was to create a more engaging and intuitive way for users to interact with the brand, moving beyond static content and traditional chatbot experiences.

Acting as a digital representative, Lumi’s role was to introduce Retune to potential clients, explain its services, and walk users through selected projects in a natural, conversational format. This allowed visitors to explore the agency’s work and positioning more fluidly, simply by asking questions rather than navigating complex pages or presentations. Importantly, this first iteration was purely customer-facing and did not include a structured data or analytics layer. However, this limitation became one of the project’s key insights. Through the nature of the interactions, it became clear how valuable customer questions, language, and decision patterns could be if properly captured and analyzed. Lumi effectively highlighted the gap between interaction and intelligence, demonstrating the need for a complementary business analyst layer capable of transforming these conversations into actionable insights. This realization became a foundational step toward more advanced AI architectures that integrate both engagement and decision support.

DINA is Lanova’s AI-powered business intelligence agent developed for a leading F&B brand to transform how decisions are made across the organization. The project started by bringing together data from multiple sources, including sales from POS systems, social media activity, customer feedback and sentiment, and relevant market and economic signals, into one connected system.

With this foundation, DINA helps the business understand what is truly driving performance. It identifies which products, channels, and external factors have the strongest influence on sales, giving management a clear view of where to focus. Instead of relying on intuition, teams can make decisions based on evidence, from optimizing product mix and promotions to shaping pricing and growth strategies.

DINA acts as a 24/7 AI business analyst, able to answer questions, explain trends, and support decision-making in real time. It brings complex data into simple, actionable insights that leadership teams can use immediately.

In addition, DINA can evolve into a customer-facing “Drink Concierge” – engaging customers directly, recommending products based on preferences, and guiding choices in a personalized way. These interactions generate valuable patterns of customer behavior, taste, and trends, which continuously feed back into the system.

This creates a powerful loop: better customer engagement leads to richer insights, which in turn drive better decisions. Through DINA, Lanova enables businesses to connect insight, action, and experience into one continuous engine for growth.

Hema was a domain-specific chatbot designed to support understanding and interaction around Coagulopathies, within the field of Hematology. The project aimed to move beyond static educational materials by creating an interactive assistant capable of guiding users through complex medical concepts in a structured and accessible way.

At its core, Hema combined a curated medical knowledge base with conversational logic to deliver clear, step-by-step explanations of key conditions such as hemophilia, thrombosis, and platelet disorders. Rather than simply providing definitions, the chatbot engaged users through guided questioning, helping them explore symptoms, underlying mechanisms, and classification pathways. This approach enabled a more intuitive understanding of coagulopathies, particularly for students and non-specialist users.

A central feature of the system was its emphasis on controlled and reliable outputs. All responses were grounded in predefined knowledge sources, ensuring consistency and minimizing the risk of misinformation. The AI agent also incorporated early elements of decision-support logic, structuring information in a way that reflected clinical reasoning without replacing professional judgment.

Hema showed that conversational AI can structure and deliver complex hematology knowledge in a way that mirrors expert thinking. By guiding users through symptoms, condition types, and mechanisms, it transformed fragmented information into a logical, step-by-step understanding process, demonstrating the potential of domain-specific AI in specialized fields.