Getting started

Up and running in one conversation.

Bricolage is a chat window. You ask it to research something, it does the deep work, and you ask for the deliverable you need. Here is the whole flow.

Start a free trialRead the FAQ
01

Brief it

Describe the outcome you want in plain language, the way you would brief an analyst. You do not write a prompt spec. Tell it the question, the context, and what good looks like.

02

Add your documents

Drop in your data room, prior decks, spreadsheets, or contracts. Bricolage works from your own material alongside the open web, so the research is grounded in your context, not just public sources.

03

Let it research

It breaks your question into sub-questions and works them in parallel across the open web, the literature, and your files, building an intermediary document for each layer as it goes. Depending on the depth you choose, a run takes anywhere from under an hour to a few days, and you do not have to babysit it. You can watch it build in real time.

04

Ask for your deliverable

When you are ready for the final output, ask for a report, a presentation, or a brief. Every claim is cited to its source, and the whole run carries a verifiable audit trail you can hand to a client, a board, or a regulator.

A Bricolage research run building in the Mind view, with thoughts, facts, matrices, deliverables and pruned dead-ends counting up
A research run in progress. Each point is a thought, a gathered fact, or a document Bricolage built along the way, and every one traces back to its source.

Still have questions?

How long a run takes, what sources it uses, how your data is handled, and how to get started, all answered in the FAQ.

Read the FAQ