Below you will find pages that utilize the taxonomy term “Ai”
How We Turned a Prototype into a Useful Market Map (with Codex)
We’ve been quietly reshaping Fresh Local Harvest over the past few weeks. The goal was simple: make it easier to find nearby farmers markets—without asking you to learn a new interface or guess the right search terms. Along the way we moved our workflow into Codex inside VS Code, which turned out to be exactly the nudge we needed to finish the job.
The short version
- We removed a fragile detour that depended on a flaky government API and committed to a single, reliable Excel → website path.
- We enriched the data so the map understands cities, states, and ZIP codes the way people actually search.
- The map now shows results near you, and pins display useful details (name, address, programs like SNAP/WIC).
- It all deploys cleanly to Cloudflare Pages when we push changes.
Why we changed course
Earlier, the project tried to do two things at once: fetch live data from a government API and process spreadsheets locally. That split left both halves half-done. We decided to focus on one path—the local spreadsheet method—because it’s dependable and gives us control over data quality.
Why Fresh Local Harvest (and Who It’s For)
Fresh food is a local story. It’s neighbors with tables, growers with early mornings, and a quick “what’s good today?” across a pile of tomatoes. Fresh Local Harvest exists to make that connection easier—and to show that you don’t need to be a programmer to build something useful with AI at your side.
Why we’re doing this
Because finding reliable, current info about farmers markets shouldn’t require detective work. We wanted a place that’s fast, clean, and up-to-date—just the essentials: what’s nearby, when it runs, how to reach them, and whether they accept SNAP.