Enterprise · Personal
File Bot
Find the file by describing it. File Bot indexes what your documents actually contain — locally, across every device, with nothing uploaded anywhere.
Overview
Filename search fails the moment you cannot remember the filename. The invoice is `scan_0043.pdf`, the contract is in a folder named `stuff`, and the photo of the whiteboard is somewhere in a camera roll of nine thousand images.
File Bot reads the contents — documents, spreadsheets, scans, images — and builds a semantic index of what each file is actually about. You then ask for it the way you would describe it to a colleague: "the shipping contract we signed last spring", "the whiteboard photo with the migration diagram".
The entire index is built and queried on your own hardware. Nothing is uploaded. That is what makes it deployable in an environment where a cloud-indexing service would be immediately disqualified.
Capabilities
What File Bot does
Search by meaning
Describe the document instead of naming it. Semantic retrieval over content, not a filename substring match.
Reads what is inside
Text documents, spreadsheets, PDFs, scans and images — including text recognised inside photographs and scanned pages.
Every device, one index
Phones, laptops and desktops indexed together, so it does not matter which machine the file landed on.
Nothing leaves the device
Indexing and inference are local. There is no upload step, because there is nowhere to upload to.
Natural-language questions
Ask what a set of documents says, not just where they are. Answers cite the files they came from.
Modest hardware
Tuned to run on the machines people already own rather than requiring dedicated GPU infrastructure.
Who it is for
Built for these situations
Enterprises with sensitive files
Teams that need semantic document search but cannot place their document store in a third-party cloud index.
Professional practices
Firms with years of accumulated case, client or project files that are effectively unsearchable in practice.
Households and older adults
Finding a document or photo by describing it removes the need to remember folder structures entirely.
Deployment
- Locally hosted — desktop, workstation or an on-premise server.
- Cross-device sync of the index across phones, laptops and desktops.
- No cloud component and no external calls at query time.
- Available as a customised enterprise deployment.
Questions
Frequently asked
Is anything uploaded to a cloud service?
No. Indexing and search both run on your own hardware. File Bot has no cloud component.
What file types can it index?
Text documents, spreadsheets, PDFs, scans and images, including text recognised inside photographs and scanned pages.
How does it work across several devices?
The index is synced between your own devices, so a file that landed on your phone is findable from your laptop.
Does it need a powerful GPU?
No. It is tuned to run on ordinary consumer and business hardware.
Interested in File Bot?
Every deployment is customised to the organisation running it. Tell us about your environment and constraints, and we will tell you honestly whether this is the right fit.