Definition
What is field-first meeting memory?
A detailed definition of field-first meeting memory and how Cairn approaches capture, projects, AI, and reports.
Definition
Field-first meeting memory is software for capturing important real-world conversations, visits, inspections, walkthroughs, client calls, and operations work, then organizing that source material so it can be searched, questioned, summarized, reported on, or turned into action later.
The field-first part matters because the work does not begin inside a neat document. It begins in a room, on a site, in a property, during a visit, or on a call where the professional may not have time to write a perfect note.
How Cairn approaches it
The Cairn loop is simple: record the work, save the memory, ask what happened later. A recording or uploaded file becomes a transcript, the transcript is reviewed and saved inside the right project, and that project becomes the place where reports, follow-ups, todos, calendar items, and AI answers can start from real context.
Cairn adds memory scope so AI can work with the context the user chooses: no saved memory, one transcript, one project, or broader saved work.
What makes it different from transcription
Transcription converts media into text. Field-first meeting memory turns that text into a durable work context. The project layer is the difference between a transcript you might search once and a body of memory you can return to over time.
Important boundaries
Cairn is local-first for core app memory: projects, sessions, transcripts, settings, chat history, processing jobs, reports, todos, and calendar events are stored locally on the user device. Some requested features still use cloud and AI services: Supabase for authentication, agent/account/billing records, Groq and DeepSeek for AI processing, and Razorpay for checkout where paid plans are available.
Cairn is built for professional work, so the product has to be honest about consent and review. Users are responsible for having permission to record, upload, transcribe, or process content, and AI-generated notes, reports, todos, calendar items, and recommendations should be reviewed before being used professionally.