Work on the money question people actually ask.

You walk into restaurants and sign them up. We hand you the list: who to call, who to visit, what they charge, and why a deal there would work. You handle the conversation and the close.

• It costs the restaurant nothing, there is no contract, and we never touch their transaction — which makes this an easier conversation than most field sales.
• You will talk to owners, not corporate. Independents only; chains are a different job on a different timeline.
• Comfortable walking in cold, in Spanish or English, and reading whether the person in front of you can actually say yes.
• You need a car and the Inland Empire in your bones. Knowing which strip mall on Base Line is worth stopping at beats any CRM.

A yes does not need a follow-up visit. An owner can claim their restaurant on a phone while you are standing there, so the close and the setup are the same conversation.

The app is React and plain JavaScript on Supabase, deliberately light on dependencies. The interesting problems are not the framework:

• Affordability maths that has to be right. Someone makes a real decision about dinner on the number we show them, so “close enough” is not a thing here.
• Menu data that is never clean. Thousands of restaurants, prices that moved, listings for places that closed — and telling those apart.
• A portal used on a phone between lunch rushes. The owner has about ninety seconds and one hand free, which is a harder design constraint than it sounds.
• Few dependencies, on purpose. If you like a stack small enough to hold in your head, that is the trade we have already made.

You would touch all of it. There is no team boundary to hide behind yet, which is either the appeal or the dealbreaker.

Every restaurant we show needs a priced, current menu. Part data work, part detective work:

• Scraping and reconciling. What a place lists, what it actually charges now, and which of those we can stand behind.
• Catching stale data before a user does. A wrong price is worse than no price — it sends someone somewhere they cannot afford.
• Closed, or just unreachable? Those look identical in a database and could not be more different in real life. Knowing which is which is most of the job.
• Comfort with SQL and a spreadsheet matters more than any particular tool.

Growth, design, support, or a job we have not thought to name. If you can point at something we are doing badly and say how you would fix it, that is a better application than a resume.

Pick “Something else” on the form below and use the last question to make the case. Be specific about the thing you noticed — that is the part we read closely.