i.
Survey
Before running anything, agents get to know your sources: the schemas, how tables relate, and how your business is shaped. They learn it once and reuse it.
For the pioneers.
Mercury Landing is where AI agents explore your data, build findings, and put together dashboards. You point them at the questions; they do the work.
Under the hood, it's a governed analytics platform. Agents query the data sources you connect, draft findings you can review, and build dashboards from them. Every statement is checked before it runs, limited to what each user is allowed to see, and logged.
Claude, ChatGPT, Cursor, or any MCP client. Connects to Snowflake, Databricks, BigQuery, and Postgres.
How it works
Most data tools assume a person is doing the work: writing each query, building each chart, laying out each dashboard by hand. Mercury Landing flips that around. You ask the questions, agents do the work, and the results come back to you as analyses, models, and dashboards. You stay involved without doing everything yourself.
ii.
Build
You don't have to file a request and wait. Mercury suggests analyses, drafts them, and refines them around the questions you actually care about.
iii.
Assemble
Nobody has to lay out a dashboard by hand. Mercury pulls findings together into pages your team can read at a glance.
Mercury Landing is for the pioneers: the teams doing work that didn't exist a year ago, the analysts who'd rather direct the work than grind through it, and anyone who's been waiting for the tools to catch up.
Tell us what you're
trying to build.
We're bringing on a small number of teams right now. Tell us a little about what you're working on and we'll set up a time to talk.