Desktop app · AI copilot

Hintify

A desktop interview and meeting copilot that listens on your computer and brings up your own stories as points you can speak from, in seconds.

Hintify landing page: Tell your best story, every time they ask
0audio uploaded
4AI providers supported
60free answers a day

The challenge

In interviews and meetings it’s easy to draw a blank on the right example. Generic AI answers don’t help, because the strongest answer is always a story only you can tell.

The tool also had to stay out of the way: no bot joining the call, nothing new in the participant list, and no audio leaving the machine.

The approach

Hintify transcribes the conversation on device, then pairs each question with your saved résumé, the job description and three to five STAR stories. A fast model streams an opener and story beats word by word, so the first words arrive before the rest is done.

After the session, a review summarizes the transcript and the suggestions you used so you can reflect and refine.

  • Local transcriptionSpeech becomes text on your machine with an on-device model; audio is never uploaded.
  • Personal story contextYour résumé, the job description and saved stories travel with every question.
  • Streaming answersAn opener you can say out loud, then the beats of your story, streamed as they’re written.
  • Bring your own modelHintify AI is included, or connect Gemini, OpenAI, OpenRouter or a local Ollama model.

From your stories to your answer

  1. 01Add storiesRésumé, job description and 3–5 stories
  2. 02Hear the questionTranscribed on your computer
  3. 03Story comes upAn opener and the beats to hit
  4. 04ReviewA report when the session ends

Built around your own stories.

The overlay shows the live question, an opener and three beats from the story that fits, along with which saved story the answer was built on.

Hintify overlay answering an interview question with an opener and three story beats

Stays on your side of the glass.

Hintify asks macOS and Windows to keep its window out of screen shares and recordings, and publishes where that protection has been tested, failures included.

Side-by-side of the user's screen with the Hintify overlay and the interviewer's view without it

Answers start while you’re still nodding.

On-device speech, preloaded context and a fast streaming model keep the time from the end of the question to the first word on screen short.

Four steps explaining Hintify's speed: local transcript, streaming answers, preloaded stories, short answers

Tools

  • Electron
  • On-device speech-to-text
  • Gemini
  • OpenAI
  • OpenRouter
  • Ollama

Screenshots captured from hintify.app.

Next projectStarbucks Customer Behavior AnalysisJupyter notebook with Starbucks customer income and spending visualizations