Gia

Project Motivation

Meal planning can be overwhelming, especially when you are juggling dietary preferences, budget constraints, and a desire to eat healthier. As someone who loves cooking, I have stood in front of the fridge many times, unsure of how to fit everything together in a balanced, cost-effective way. Through observations and conversations with friends and family, I realized I was not alone in this struggle. Whether it is a busy parent trying to manage family dinners or a young professional crunched for time, many people find meal planning more stressful than helpful.

Those challenges led me to develop GIA, a generative AI meal planning assistant integrated with the Whole Foods app. GIA curates recipes based on dietary preferences, health goals, and budget parameters, then translates them into organized grocery lists that reference real time pricing and inventory. It also scales recipes with ease, accommodates allergy friendly ingredient swaps, and offers cost conscious alternatives. Motivated by the desire to go beyond basic recipe apps, I saw artificial intelligence as a powerful tool to help more people enjoy cooking while streamlining everyday decisions around food. GIA’s ultimate purpose is to reduce the stress of meal planning, leaving individuals with more time for the people and activities they value most.

‍

My Role and Goals

I served in several capacities to ensure GIA would address genuine user needs and remain technically robust. As Project Manager, I designed the timeline and assigned responsibilities to balance creative brainstorming with clear deliverables, maintaining a consistent vision throughout the project. As Conceptual Lead, I channeled my own experiences with meal planning into brainstorming sessions and important financial and technical decisions. As Research Coordinator, I conducted interviews, analyzed market research, and tested prototypes with peers to confirm the importance of convenience, budget awareness, and personal health. Above all, I aimed to demonstrate how generative AI can support real life tasks. Rather than treating AI as a novelty, I wanted GIA to provide tangible benefits, help individuals adopt healthier habits, and adapt to varied financial and dietary requirements.

‍

Scope

From the outset, this was a rapid ideation and proof of concept rather than a fully developed product launch. We took a Wizard of Oz approach that used Figma screens to simulate GIA’s core capabilities including recipe generation, grocery list creation, and cost calculations before any production level coding began. By benchmarking these capabilities against large language models such as Perplexity AI, we confirmed that constraints involving allergies, budgets, and portion sizes could be managed effectively.

‍

Our Process: From Ideation to Implementation

We started our semester with an expansive matchmaking and ranking brainstorming session that included everything from AI-driven wardrobe assistants to text-to-video chatbots.

To narrow down these ideas, we used a “Drunk Island” test, which helped us quickly judge whether a concept was feasible with current AI capabilities or if it demanded near-superhuman intelligence or accuracy. Meal planning emerged as a standout option because it matches what large language models already do well, and it addresses a challenge people face on a regular basis.

To evaluate the broader impacts of a GenAI meal planning app, we conducted a Consequence Scanning exercise. This allowed us to identify positive outcomes such as enhanced dietary awareness, greater culinary experimentation, increased sustainability awareness, and valuable data for research labs. At the same time, we acknowledged potential risks around biased recommendations or cultural insensitivity, highlighting the need for careful oversight and proactive measures to safeguard the user experience.

Continuing our research, we interviewed single parents, students, and working professionals to better understand their daily cooking habits and budgeting concerns. These conversations revealed how the gap between online recipes and actual grocery lists often leads to stress and inefficiency. Recognizing the growth of online grocery services, we envisioned GIA as a feature within Amazon Prime and Whole Foods, bridging recipe discovery with real-time inventory and pricing data.

As part of our iterative design, we tested GIA’s core capabilities by prompting the AI to handle real-world scenarios such as “Plan a week of gluten-free, budget-conscious dinners for two.” We used Perplexity AI to validate the system’s ability to manage dietary constraints, cost ceilings, and portion sizes. Through these tests, we confirmed that large language models can effectively address the varied and evolving demands of meal planning.

We then added features like real-time grocery list generation, cost-saving ingredient alternatives, and adaptive personalization that learned from user behavior over time.

‍

PERSONAL REFLECTION

As the person who initially conceived the idea, I wanted to ensure that GIA was designed to solve a real problem, not just serve as a justification for AI implementation. Too often, AI-driven projects prioritize technical capabilities over genuine user needs, treating AI as the solution rather than a tool to enhance human experiences. Throughout this process, I focused on testing GIA’s real-world usefulness by running queries on Perplexity, gathering feedback from friends, and refining our approach based on actual user challenges. This experience reinforced that AI should not be integrated for its own sake but should be thoughtfully applied where it can create meaningful, tangible benefits. GIA is not about proving what AI can do; it is about ensuring AI serves a clear, practical purpose in making meal planning more intuitive, affordable, and accessible.

‍

TEAM INSIGHTS

Our biggest challenge was proving that generative AI could consistently handle detailed meal-planning prompts without tripping up on allergies, dietary restrictions, or cost constraints. Initial skepticism led us to refine our demos and test edge cases more thoroughly. We also solidified the idea that GIA should be a B2B partnership with Whole Foods rather than a separate consumer-facing app, which clarified how to integrate real-time data and develop a revenue model.

By the end of the project, we had a compelling proof of concept: a solution that addresses a universal pain point, delivers tangible value to a major retail partner, and maintains a focus on user well-being.

‍

CONCLUSION

GIA represents a forward-thinking approach to meal planning, using generative AI to reduce decision fatigue and make grocery shopping more intuitive. When I started this project, I was not interested in creating just another AI-powered tool. I wanted to design something that felt genuinely useful, something that could help people like me who have struggled to balance health, budget, and variety in their meals. By integrating GIA into an existing ecosystem like Amazon Prime and Whole Foods, it becomes more than just a convenient feature. It becomes a practical, scalable solution that makes meal planning easier without adding more complexity. From the earliest brainstorming sessions to financial modeling and ethical considerations, every step of this project was guided by a focus on real user needs. Seeing GIA take shape reinforced my belief that AI should work in service of people, not the other way around. I am excited about the potential impact of this project and how it could reshape the way people approach cooking and grocery shopping.