OpenInterviewer · Open source
From the scripted demo

Reconstructing purpose becomes part of the cost of reading, so the saved item feels like unfinished administrative work. “I had forgotten which project it was for, so opening it felt like work before the reading even started.”Scripted demo · Maya · turn 4

Follow the answer, not just the script.

Turn a research guide into an adaptive interview. Participants get thoughtful follow-ups; researchers get interpretations linked back to transcript evidence.

Try the scripted demo · 2 minSelf-host your own
Safe to try immediately Fictional participant, fixed branches, no account, API key, live AI, interview API call, or saved data.
The research loop

From a question to evidence you can inspect.

  1. 01

    Frame the study

    Define a research question, topic guide, and the context you need from participants.

  2. 02

    Follow the thread

    Probe what a participant means instead of mechanically asking the next prepared question.

  3. 03

    Trace the insight

    Review interpretations, nuances, and hypotheses alongside their transcript evidence.

Configured researcher workspace

Sign in to your workspace. Hosted mode guides you through connecting your own provider and storage; standalone uses the operator's configuration.

Researcher sign in

Self-host

Deploy the MIT-licensed application with your own provider keys and Upstash database.

View deployment guide