Startup Memorial

Every failure is a lesson in disguise. Decode the past to engineer the future.

Comment

Inspiration

As a fellow hackers, we sometimes meet the struggle of finding idea or even get stuck for a long time when brainstorming ideas. We believe if we have more experience in brainstorming ideas, we can expand our perspective and explore a lot of other solutions of a problem that we may not think about right now.

What it does

Startup Memorial is a web app that archives startup failures and turns them into learning cases. Users can choose a case and read the description of the startups with some AI-generated analysis. Additionally, users can ask more questions about the case from provided AI chatbot. Then, users can propose a solution and get a feedback from the AI including a score 0-100. Other users can upvote or replies to the ideas submitted. Users can also add a new case with AI help to fill in some missing information.

How we built it

We built Startup Memorial as a React + Typescript web application with Firebase Firestore as the backend. We allow real-time updates and scalable discussion threads by storing each startup as a document with proposed solutions stored as subcollections, and replies as subcollections of solutions. We generate the initial frontend using Google AI studio where the structure is component-based. The styling we used is Tailwind CSS. We modified it to listens to Firestore so it updates instantly every changes made without refresh. We integrated an AI model (Gemini) to generate startup failures analysis, giving feedback to solutions, and also fill in some accessory information.

Challenges we ran into

One of the main challenges was designing a Firestore data model that could support startups, solutions, replies without becoming deeply nested or hard to scale. We initially tried to store replies directly insider solution objects, which caused state and synchronization issues. We solved this by restructuring the data as subcollections and relying on real-time listeners instead of manual state updates.

Accomplishments that we're proud of

We're proud of the idea that can help a lot of people to improve their skills to brainstorm new and creative ideas. We are also proud that we are able to integrate AI to our project.

What we learned

With the presence of AI, we con focus more on insight and decision-making instead of struggling on building everything from scratch. Also, we can use AI to summarize and generate the analysis of the data rather than creating raw data processing logic ourselves. Thus, AI can design a better story-telling of the data automatically.

What's next for Startup Memorial

There are a lot of basic features that can be added such as users account and security. Additionally, we can make some feature extensions such as Streaks, What-If Scenarios, and Bookmarks/My Library.

Built With

Share this project:

Updates