Bridging food discovery and creativity by combining an AI-powered cultural food chat assistant with intelligent recipe generation that transforms random pantry items into delicious, data-driven meals.
Role: Lead Product Designer(UI/UX)
Timeline: 3 weeks
Platform: iOS & Android
Setting the Scene
"I open my fridge and just see random things… then I waste them.” — Interviewee, 31
Users weren't wasting food because they didn't care, they wasted it because they couldn't connect ingredients into meals. Tom, a busy professional worker
At the same time, many users wanted a place to ask simple questions about Nigerian foods without scrolling through endless blogs and recipe sites.
The Challenge
How might we:
Help users instantly transform pantry chaos into recipes?
Reduce food waste by suggesting flexible substitutions?
Make AI food help feel personal, like chatting with a friend?
Existing solutions often lacked cultural context and flexibility:
Recipe sites required long searching.
Apps didn't adapt to what ingredients users had.
No direct way to ask questions about local dishes
My Approach
I combined an AI Recipe Generator with LuChat, an interactive food chatbot.
1. Framing the Solution
Chat with Lu to ask questions about Nigerian meals or pantry items.
Generate instant recipes from typed ingredients.
Offer substitutions when something is missing.
2. Design Execution
Conversational UI: lightweight, approachable, and culturally tuned.
Recipe Cards: show prep time, macros, substitutions upfront.
Regenerate FAB: a floating button for exploring endless variations.
3. Collaboration
Worked with AI engineers on ingredient parsing + Nigerian food NLP.
Partnered with developers to ensure fast, low-latency results.
The Breakthrough
When Tom scanned his fridge and saw spinach turned into a creative recipe, he said:"I never thought of using spinach this way, this saves me a Google spiral." That was the aha moment: LuChat wasn't just AI, it was a kitchen companion that turned waste into inspiration.
My Process
Step 01
Discovery & Research
These findings shaped the synergy between LuChat and the AI Recipe Generator. Users' craving for natural, intelligent food conversations inspired LuChat’s tone and memory features, while their frustration with wasted ingredients and decision fatigue led to a pantry-based recipe engine that makes everyday cooking feel effortless, personal, and fun.
Step 02
Concept & wirefreames
Our early wireframes explored minimal chat flows with lightweight recipe suggestions. As the system matured, we integrated pantry-based logic, real-time ingredient matching, and inline recipe previews. These additions deepened functionality while keeping the LuChat experience effortless and human.
Step 03
Final Design
Final designs brought LuChat to life with a clean conversational interface, dynamic AI responses, and real-time recipe generation from pantry data. Every interaction feels personal and purposeful, blending warmth, intelligence, and clarity in one seamless experience.
“I love that I don't have to think. The card just tells me what's good for me today.”
Outcomes
80%
Reported less frustration choosing meals
50%
Spent less time deciding
+30%
Increase in homepage interactions
During usability testing, 80% of participants reported feeling less frustration when choosing meals, while 50% spent less time deciding. These results validated our hypothesis: when choices are clear and contextual, users feel more in control of their nutrition.
Future Iteration
Future iterations could allow users to order meals directly from the daily recommendations, while the grocery list could evolve into a checkout-ready shopping experience.
