Mason Bee

Link for presentation:

https://www.canva.com/design/DAHDRQJDHCQ/Vr7D9tLlX_YAJnxyEoTC7A/edit?utm_content=DAHDRQJDHCQ&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton

Framing the Problem

What is the problem you are trying to solve? Who does it affect?

The Problem

• Millions of pieces of good furniture end up in landfill every year, not because they're broken, but because owners couldn't find a buyer.

• In the UK, only 17% of discarded furniture is ever recycled. The rest? Dumped.

• Rising costs push consumers toward cheap, synthetic furniture that is hard to repair and impossible to recycle, creating a throwaway cycle.

• Inflation has stretched household budgets to breaking point, making brand new furniture out of reach for many.

Who It's For

• First-time buyers furnishing a flat on a budget.

• Families looking for furniture built to last.

• Renters who want quality without the price tag.

• Contractors who want a platform for local jobs.

• Clients who need reputable local contractors.

Idea Explanation

What is your idea? How does it fix the problem?

Mason Bee is a platform that gives furniture a second life, and gives people a smarter, more sustainable way to furnish their homes.

The Platform

• A centralized app for DIYers, renovators, and anyone furnishing a home on a budget.

• Seamlessly connects buyers with quality second-hand furniture and verified tradespeople in one place.

• Removes the friction that causes good furniture to be thrown away in the first place.

Smart Budgeting

• Real-time budget checker keeps you on track throughout your project.

• AI recommends furniture and layouts based on your budget and preferences.

Verified Tradespeople

• Access profiles of reputable, company-recommended contractors and tradespeople.

• Languages listed on each profile so clients can choose someone they're comfortable with.

• Product and builder reviews for safety and reliability.

AI-Powered Personalization

• Analyses user preferences and behavior to suggest furniture, layout styles, and color schemes.

• Real-time AI chatbot identifies what you're looking for as you browse.

• Recommendation engine learns from your interactions to surface the most relevant options.

Community

• Users can share designs in a community gallery for feedback and inspiration.

• Encourages exploration of different styles and approaches.

Implementation

• How do all the pieces fit together? Does your frontend make requests to your backend? Where does your database fit in?

• The website mainly works as a frontend application built with React.

• The frontend manages all user interactions, such as the quiz, chatbot, and viewing recommendations.

• When AI is needed, the frontend sends a request directly to Google’s Gemini API.

• Gemini returns a response, which is stored temporarily in the app and displayed to the user.

• There is no backend server or database, so any user data is not saved permanently and resets on refresh.

Challenges

What did you struggle with? How did you overcome it?

Development Challenges

• Manual logo implementation required locating and updating each file individually due to AI misplacement.

• Connected to external databases to overcome the challenge of launching with no existing content since we are a community based app.

Performance & Speed

• Optimized AI inference to minimize latency and deliver real-time recommendations instantly.

• Background processing handles AI tasks without interrupting the user experience.

Accomplishments

What did you learn? What did you accomplish?

Technical skills

• We learned how to use Google AI Studio and explore its capabilities.

• We improved our ability to write effective prompts to get better outputs from AI tools.

• We learned how to navigate GitHub, create repositories, and organize our project files.

Teamwork

• We learned how to collaborate effectively and share ideas as a team.

• We practiced allocating tasks and being flexible with our roles depending on what the project needed.

• We supported each other and adapted to unexpected challenges during the process.

Next Steps

What are the next steps for your project? How can you improve it?

Monetization

• Earn a percentage of the commission when contractors get hired through MasonBee.

• Create a subscription based model to unlock premium features such as styling tutorials. 

• Creating paid enterprise accounts for interior designers to access unique furniture options while also receiving discounts. 

User Personalization

• Allowing users to create profiles based on their roles, so that the platform can provide them with more relevant features and recommendations. 

• Develop machine learning models that learn from user behavior and design trends to improve furniture recommendations and personalization.

• Add AR room-scanning tools that allow users to scan their room with their phone camera, measure dimensions accurately, and visualize furniture placement in their space before purchasing.

• Create interactive tutorials with visual guidance so users can learn how to renovate parts of their home and assemble their furniture. 

• Notifications which allow users to be notified when a desired style of furniture is available.