Build my first project: Sushi Preference Tracker

What: I want to create a software solution for Sushi businesses that helps them predict what sushi people like and dont like, so they can waste less food and sell more.

When: The idea would be implemented on top of existing systems that exist for sushi business owners and sushi trains.

Where: Would give it away for free to test the data locally, then would aim to sell it world wide and starting in English speaking countries and Vietnamese.

Why: It will save them money, it will allow to sell more because more options will be available to those who want those options.

How: After a couple months they would have more cash flow, less waste and more customers coming back for their favourites.
The project I chose: Sushi Preference Tracker
What I want to prove: The should be able to reason over the data of old sales if it exists, if it doesn't exist it should replace their typical ordering system, so that the new data with the right measurements can then be measured. So the user should able to create a menu, then use it to process orders. The Focus should be on that customers can self order as well.

Interview me before building
Read my original interview, chosen project, business rules and acceptance checks. Review the report, app files and examples I have supplied. Start from what you already know about my business and carry my decisions forward. Open our conversation with the business or design decision that would most improve the result. Handle installation and publishing when we reach those steps.
Interview me to get everything you need to build the best result for my business. Explore who will use it, what they need to accomplish, what happens in practice, which decisions and exceptions matter, how the app should look and feel, and what a successful outcome looks like. Explore data, permissions, devices and integrations where they affect the work.
Follow my answers wherever they reveal something important. Ask for concrete examples, discuss trade-offs in my language, and resolve gaps or conflicting requirements with me. Continue the interview for as many rounds as the project needs. Use fictional or redacted examples when discussing actual work.
Once you understand the outcome, summarize how the app will work in a representative case and what will count as success. Carry my answers into the design, business rules and acceptance checks, then build and test that workflow.

Getting started
I am starting from scratch. If this workspace can build and run the app, start here. If setup is needed, ask which device I use if you cannot tell, and guide me one step at a time.

Build instructions
To implement the Sushi Preference Tracker, start by designing the first screen around a sushi menu object. This object should have fields for dish name, price, and popularity, and actions to add, remove, and edit dishes. Use the approved brand colour #F204F6 and create a typographic identity for the sample business. The user of this object is the sushi business owner, who will use it to create a menu and process orders. The next action that changes the state of this object is the addition of a new dish to the menu. For example, if the business owner wants to add a new dish called 'salmon sushi' with a price of $10.99, they should be able to do so and see the updated menu. The system should also allow customers to self-order, so include an action for this as well. The initial screen should be simple and easy to use, with all inputs and actions within reach at a 320 CSS-pixel viewport. Use grid and flex children to make the layout responsive, and include a second working view for when there is a genuine handoff or saved record to reopen. In terms of technology, the system will be implemented on top of existing systems for sushi business owners and sushi trains. The design should be accessible and follow the principles of inclusivity, with clear and consistent typography, spacing, materials, and imagery that suit the trade. To test the system, try entering a week's worth of sushi sales data and see which types of sushi are most popular and which are least popular. Also, test the system with a new sushi business with no existing sales data, an existing sushi business with sales data, and a customer who wants to self-order. The system should be able to identify a significant reduction in food waste after using the tracker for a month, and the business owner should be able to create a menu and process orders successfully, with a focus on customer self-ordering.

Workflow examples that must pass
new sushi business with no existing sales data → add dish to menu (stage = Before launch, dish_name = salmon sushi, price = 10.99)
existing sushi business with sales data → process order successfully (stage = After launch, dish_name = salmon sushi, price = 10.99, popularity = high)
customer wants to self-order → allow customer to self-order successfully (stage = After launch, dish_name = salmon sushi, price = 10.99)

Design
My chosen brand colour is #F204F6. Use it consistently as an accent, provide a readable foreground on every surface, and keep important status differences understandable without colour alone. design the first screen around a sushi menu object with fields for dish name, price, and popularity, and actions to add, remove, and edit dishes, using the approved brand colour #F204F6 and a typographic identity for the sample business
Design the first screen around an object this business already works with and the decision someone makes from it. Name that object, its user, its fields, and the action that changes its state. Give it visual weight and use the owner's vocabulary. Use the approved brand colour and any supplied logo; otherwise create a typographic identity for the sample business. Pick typography, spacing, materials and imagery that suit the trade. Include a second working view when there is a genuine handoff or saved record to reopen. At a 320 CSS-pixel viewport, keep every input and action within reach. Let grid and flex children shrink and wrap. Give a wide table or drawing its own scroll area. Check the result where a phone user changes the value. Keep engineering checks, the sales pitch and build instructions outside the operator interface.

Recommended technology
- existing systems for sushi business owners and sushi trains

Acceptance checks
- new sushi business with no existing sales data → add dish to menu
- existing sushi business with sales data → process order successfully
- customer wants to self-order → allow customer to self-order successfully
- identify a significant reduction in food waste after using the tracker for a month
- create a menu and process orders successfully
- allow customer self-ordering successfully

Preview and handoff
Use fictional sample records and let me try the complete first task. Explain what you actually tested and what remains to verify. Give me the source files and show me how to reopen the app locally. Then offer the Cloudflare publishing steps below.

From the demo to my application
After I can try the demo, tell me which checks passed and what still needs testing. Ask me: “Would you like me to build this application from this demo?” Wait for my answer.
If I reply “Yes, build this application”, or clearly agree, start implementing the first usable version in this coding workspace. Carry forward the approved design, workflow, source files and acceptance checks. Keep the demo available as a reference. If implementation reveals a gap, return to the interview and work through it with me.
My yes authorizes building and testing. Get separate approval before publishing, enabling charges, connecting live accounts or customer data, or sending real messages.

Put my finished app online with Cloudflare
Once the first app passes its checks, offer to publish this exact version at a free workers.dev address I can open on my phone or share. Tell me the link will be public, show the proposed project name and sample records, and ask whether I want to publish now. A local build can stay local if I decline.
Use my own Cloudflare account. If I am new, guide me through creating a free account at https://dash.cloudflare.com/, verifying my email, and choosing my account subdomain in Workers & Pages. If I already have an account, reuse it. Let me handle sign-in in my browser without pasting a password, token or one-time code into chat. If using the CLI, use npx wrangler login as needed and npx wrangler whoami to confirm the intended account.
Package only the public app assets in a clean folder with index.html at its root. Use a new, available Worker name and a Wrangler configuration with today's compatibility date, workers_dev set to true, and assets.directory pointing to that folder. Keep the master prompt, report, private records and credentials outside the assets directory. Use Workers Free for this static app. If setup requests payment or would replace an existing app, pause and explain. A custom domain is optional.
After I approve, run npx wrangler deploy in the configured project. If this environment cannot run it, guide me through the Workers dashboard static file upload one step at a time. Copy the actual workers.dev URL from Cloudflare's deployment output. Open the link in a fresh browser session, reload it, and try the main acceptance case. Give me the verified link and tell me how to publish updates to the same Worker. Use the official guide if the dashboard has changed: https://developers.cloudflare.com/workers/static-assets/get-started/. If you cannot access the link, say what remains for me to check. Keep live integrations and paid services separately opt-in.