← Back to the foundation
ALPHA LENS IN DEVELOPMENT

Understand before
AI sees it.

Useful context starts with better preparation. Alpha Lens is being built to turn documents and raw information into clear, structured input — with deliberate control over what an AI receives.

See how it works What will the free demo include? ↗
ALGORITHM FIRSTLOCAL FIRSTHUMAN CONTROL
01 / THE PURPOSE

Less noise.
More useful context.

A document is more than a block of text. Structure, repeated content and sensitive details all affect what happens next.

01

Make information usable

Extract readable content and preserve useful structure so downstream tools have something clearer to work with.

02

Prepare before prompting

Divide longer material into manageable sections and prepare compact context for a chosen task.

03

Review what leaves

The privacy direction is to identify supported sensitive-data formats and let a person review the prepared text before sharing it.

02 / HOW IT WORKS

A preparation layer.
Before the model.

Run practical processing first. Use AI afterwards, when interpretation adds value.

  1. 1

    Provide the source

    Choose text or, in a future supported runtime, a document or structured dataset.

  2. 2

    Prepare the content

    Extract, clean and organise the information with task-specific processing.

  3. 3

    Review the result

    Inspect the prepared text and supported privacy findings before using it elsewhere.

  4. 4

    Choose the next step

    Keep the output locally or explicitly pass selected context to your own workflow.

ALPHA LENS / TEXT PREPARATIONILLUSTRATIVE EXAMPLE
SOURCE TEXT
Supplier follow-up

Please contact jane@example.com
or +44 7700 900123 about the delivery.
The revised date is 22 October.
REVIEWABLE OUTPUT
Supplier follow-up

Please contact [EMAIL_01]
or [PHONE_01] about the delivery.
The revised date is 22 October.

A planned preview of supported replacements. It does not establish that a document is anonymous or safe to share.

03 / WHERE IT FITS

One layer.
Several useful paths.

DOCUMENT WORKFLOWS

From a source to readable context

The code foundation includes document extraction and OCR modules. These need further validation before public file processing is offered.

KNOWLEDGE & RETRIEVAL

Give long material some structure

Text splitting and summary preparation form part of the intended workflow for research, retrieval and focused analysis.

BUSINESS INFORMATION

Prepare data before interpretation

Data and invoice modules provide a foundation for extracting useful fields and measurements. They are outside the first public demo.

PRIVACY WORKFLOWS

Start with a reviewable text preview

The first demo is planned around limited contact-data detection. Broader entity detection and anonymisation are separate development work.

04 / THE FIRST PUBLIC STEPPLANNED · NOT YET AVAILABLE

A small window
into Alpha Lens.

Try a short piece of text, see supported findings and review the changed output. A focused introduction to one part of the product.

What you will be able to try

  • Paste up to 3,000 characters of test text.
  • Clean up excessive spacing and blank lines.
  • Find email addresses and selected phone-number formats.
  • Review replacements alongside the original text.
  • Load a fictional sample, clear the input and copy the result.

What stays outside this demo

  • Document uploads, OCR and image processing.
  • Name recognition or comprehensive PII detection.
  • Batch processing, integrations and API access.
  • The production engine, private rule library and configuration.
  • AI-generated answers or automatic external actions.
Planned processing: in your browser. The demo is intended to use a separate, limited set of public rules without submitting the pasted text to a server. Until it launches, this page provides product information and illustrative examples only.

Supported formats can still be missed or incorrectly matched. You must review the output; a clean preview does not guarantee removal of all sensitive information.

05 / BUILT PIECE BY PIECE

The wider product
comes after validation.

The longer-term direction includes a licensed local runtime and documented API or MCP access. Availability will follow working implementations and testing. The public demo will not expose the full engine.

Explore the Alpha IOS ecosystem