How to build a voice recipe app
The author built RecipeNinja using Windsurf, later adding Claude Code and Gemini help, with a voice assistant that navigates the app during conversation.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Build [recipe application] with a Rails API and React frontend. Add voice-to-voice interaction through [realtime API configuration] and function calls for [navigation actions]. Demonstrate voice interaction and recipe generation and editing for my review.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- tomblomfield
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author reports building a roughly 35,000-line recipe app in two to three weeks, including interactive voice assistance and function-based site navigation.
- Human involvement
- The author chose the stack and shared demos for feedback; the post does not describe code review or testing steps.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Used Windsurf to build a Rails 8 API backend and React frontend.
- Added voice-to-voice responses using the OpenAI realtime API.
- Used Claude Code and Gemini 2.5 Pro for trickier features.
The author built RecipeNinja over two to three weeks, aiming for the recipe application they had always wanted. They describe themselves as a founder turned investor with earlier PHP and Ruby on Rails experience, but no professional Ruby work since approximately 2015.
They chose Windsurf to build a Rails 8 API backend and React frontend. OpenAI's realtime API provided voice-to-voice responses. Over the final few days, they also used Claude Code and Gemini 2.5 Pro to help with trickier features. The post describes the resulting application as roughly 35,000 lines of code.
A fully interactive voice assistant lets someone use the app while cooking without touching an iPad with dirty hands. Function calling allows the assistant to navigate the site in real time during conversation. The author shared demonstrations of the voice assistant and of generating and editing recipes, and invited feedback. The post does not describe test results, deployment procedures or a code review process.
Who it's for: developers exploring a recipe application with conversational navigation and hands-free kitchen interaction.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · tomblomfield
Frequently asked questions
What stack did the author choose?
Windsurf was used to build a Rails 8 API backend and React frontend, with OpenAI’s realtime API for voice-to-voice responses.
What could the voice assistant do?
The author says it enabled hands-free interaction while cooking and used function calling to navigate the site during conversation.
More details
Remaining steps
- Shared demonstrations of the voice assistant and recipe generation.
About this account
The coding agents supported implementation while the author directed the project and supplied context.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=43553031. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: tomblomfield on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
Windsurf · Rails 8 · React · OpenAI realtime API · Claude Code · Gemini 2.5 Pro
Agentic coding · Development
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.