Building an accessible solution for collaborative cooking

Diagnosis

Context

The project was based on our direct observation in major cities like Paris: student food insecurity and the underutilization of existing spaces. Students lack adequate space, resources, tools, organization, and knowledge to cook healthy, diversified meals. Our vision was to transform unused kitchens into low-cost hubs for social exchange and practical culinary learning.

Problem

How can a digital service provide students in major cities with the necessary resources to foster healthy, consistent food habits? Thereby transforming a logistic constraint into a social opportunity?

Role

Partnering with a Data Scientist, I drove the discovery phase, from primary user research and physical proof of concept to low-fidelity prototyping. My job was to ensure the user experience came first, and that our project evolved toward an economic model serving genuine human connection.

Key insights

Initial research identified the barriers to student autonomy: not just cost and space, but critical deficits in organization, nutrition and cooking knowledge.
The service's value proposition is social engagement and learning. Users seemed highly motivated by the chance to meet, share culture, and learn new cooking skills.
The matching algorithm must be human-centered, prioritizing affinities like shared values, desire to teach/learn, and of course food habits.

Conception

Methodology

I adopted a terrain-validated approach. The project prioritized a non-digital Proof of Concept (PoC) before any interface development, to ensure the core concept was viable. This saved development resources and focused the MVP scope.

Design conception

We used the validated user journey and insights from the physical PoC to define the essential functionalities and logic of the first digital prototype. A key strategic decision was transforming the basic Host Profile into a Social Trust Passport, prioritizing features that convey safety and community feedback over simple listing details. To ensure a high adoption rate, we formalized the matching criteria (age, distance, food preferences, quantity) which served as the technical specifications for the Data Scientist to optimize the recommendation algorithm. Additionally, I built this first prototype as an MVP to ensure future testing and iterations.

Results

Delivery

Thanks to the user research synthesis, we validated the core frictions, starting with the lack of trust between strangers. Once field testing confirmed a basic user journey, we built the Figma prototype and worked toward a local MVP.

Metrics

Adoption target: 200 new user sign-ups within the first 3 months post-launch.
Trust target: maintain a profile completion rate of 80% or higher, reflecting user investment in the trust-building mechanisms.
Match rate target: exceed a 60% Host & Cook match rate within the first 3 months, validating the accuracy of the affinity-based algorithm.
Social value target: 75% of users reporting a new cooking skill or organizational method learned after their first session, validating the pedagogical component.