Open million-row files, run agents in formulas, and trace every change—so analysts can work in the grid, not around it.
1.2M-row file, formulas that run agents, attribution on every save
Ninety percent of what analysts want from AI is not automation. They want to get into the data and manipulate it on the fly. The tools that promise that either choke past a few hundred thousand rows, live in a chat window disconnected from the grid, or make changes nobody can trace.
And when a change is wrong the night before the board meeting, "the AI did it" is not an audit trail.
The grid is the interface—now with memory and accountability.
Open anything. Upload.xlsx or.csv, connect a SharePoint workbook, or pull from Databricks, Snowflake, or the lakehouse with Sheet Queries, refreshable and server-side.
Put the agent in a cell. =ATHENA for a plain-language question. =ATHENAQUERY for a governed query through your semantic model. =AOP to run a workflow per row. =TOOL for any of 600+ tools. =ATHENAASSET to embed an asset. =IMPORTRANGE across workbooks.
Transform in plain language. "Bucket these by region and quarter, format as currency, add a total row." The agent writes and runs it, in Python if needed, inside the workbook.
Trace every change. Author, human or agent. Session and message. Before and after. Compare in Editor. Roll back by range.
Studios. Word, PowerPoint, and Sheets as live documents. Full-fidelity import and export, your templates, track changes, Compare. Sheets handle 1M+ rows with =ATHENA(), =ATHENAQUERY(), =AOP() in the cell.
| Capability | People can | Agents can | Together |
|---|---|---|---|
| Studios | Edit as you always have, in the browser or through the Office add-in. | Build the deck, draft the memo, fill the model, in the same file. | Live co-editing; every change signed; accept or roll back by range. |
Anything a person can do here, an agent can do with the same permissions, and both land in the same audit trail.
=AOP("contract-review", "file", A2). Drag down 500 rows. 500 governed runs, each attributable to the person who dragged, each under a hard budget. Reference the structured result in the next column: =B2.summary, =B2.parties[0].name.
Refreshable, server-side pulls from Databricks, SQL catalogs, the lakehouse, semantic models, and SharePoint Excel. Existing Power Query workbooks keep their linkage.
Dozens of department spreadsheets in different shapes normalized into one table. A Fortune 500 retailer's two-week Finance and HR task finished in 15 to 20 minutes; the prototype was built overnight.
An agent changed 4,000 cells the night before. Open the Activity Monitor, see the range, the conversation that caused it, the before and after. Roll back row 212. Keep the other 3,999.
Block read or write on specific ranges of a specific sheet. The assumptions tab is yours; the model tab is the agent's.
Same agent formulas and Sheet Queries inside desktop Excel. "Copy to Athena" uploads the workbook; sync brings edits back.
“Gave Athena the spreadsheet, explained which column I was trying to figure out, and then ten secs it told me... exactly how the calculation was based on the other data.”
“One workspace. All different, let's say, avatars of their output data. Right? Suddenly, it's dashboard. Excel is here. PowerPoint is when it's all together. It's very impressive.”
“It's good to understand the capability of the platform. That how far, and how quickly you guys can get these things going [...] that has been impressive overall.”
“Lets us get a lot more out of our data than [...] We can sit on top of all of the data in [our data warehouse] as opposed to just, like, our little ingested, you know, three gigabytes semantic models. I think we have a lot of power here.”
“He wrote the entire stuff. By itself. And I was there, and I was thinking, alright. If I had to write all this stuff, I mean, I would've spent days, days, days.”
“I send that to the programmer and he said, alright. I read your file. I copied the section of the code because he has the right interface, and it works. Wow. Just like magic.”
“Use the Athena platform probably closer to the concept and how it was designed, like, with the spaces architecture where you've got a set of primitives and capabilities, and you bring together the right set of capabilities to solve a task.”
“Digital worker, it gets spun up, it gets given access to the toolkits that it needs. It gets given access to the data that it needs. Just exists for the period it needs to exist to perform the task. And then it gets torn back down again afterwards.”
“I like that direction, that concept of the agent is the agent. It becomes an entity in its own right with its own ownership and its own missioning, and you get rid of the gray area and the blurry lines.”
“It was PowerPoint... there was no kind of distinction between the two. So that was killer. And I literally had a meeting on it today, and people said how great it was.”
Verbatim from customer calls. Customers anonymized.
Sheets is where analysts live. Agents work in the grid ad hoc. Workflows (Agent Operating Procedures) fill templates on schedule. Applications turn a sheet into a dashboard the many can open, included in Athena services.
Attribution per save: human or agent, session, message, before and after.
Compare in Editor; selective rollback by range; version history.
Range-level permissions for agents.
Budgets on =AOP fan-out. Fails closed if the budget system is unreachable.
Attribution is per save, not per keystroke. If a human and an agent edit the same cell inside one save window, you get the last writer. The before-and-after diff still shows in Activity Monitor.
Agents can build and explain models inside it, it handles enterprise-scale data, and every change is attributed.
Yes. Sheets imports and exports Excel files.
Yes. A cell can call a workflow and return the result.
Records every change by a person or an agent, rolls back one contributor's edits without losing anyone else's, and keeps the model, instructions, and sources behind each agent action.
How the platform is deployed: Athena's managed cloud, your cloud on AWS, GCP, or Azure, on-prem, air-gapped, or GovCloud. Same platform in every option.
Build, Work, and Data on top; 150+ connectors and every surface in and out; one map of the whole platform.
We will open it, ask it a question, and show you who changed what.