Windows Data Privacy Scanner Development

Project Title:
Build a Windows Desktop Privacy & Sensitive-Data File Scanner

I need an experienced Windows software developer/team to build a professional desktop application that scans a computer deeply and identifies files containing sensitive or confidential information.

The application must work mainly locally/offline, so customer documents do not need to be uploaded to external servers.

The software must inspect the actual contents of files, not only filenames.

Core Purpose

The app should help businesses discover forgotten or exposed sensitive information such as:

- Social Insurance Numbers (SIN)
- Bank/account numbers
- Credit card numbers
- Passport/driver licence numbers
- Dates of birth
- Names, addresses, phone numbers, emails
- Health information
- Employee/customer information
- Tax-related information
- User-defined confidential keywords/patterns

Deep Scanning

The application must scan:

- Entire drives
- Selected folders
- Deeply nested folders
- Hidden files/folders
- System folders where permissions allow
- USB/external drives
- Network/shared drives where authorized

The program must never silently skip files.

Each discovered file should receive a status:

- Successfully scanned
- Access denied
- Password protected
- Encrypted
- Unsupported
- Corrupted
- Locked
- Skipped with reason

File Types

Minimum support:

- PDF
- DOC/DOCX
- XLS/XLSX
- TXT
- CSV
- RTF
- PPT/PPTX
- JPG/JPEG
- PNG
- TIFF
- BMP
- ZIP

Preferred additional support:

- EML
- MSG
- PST
- 7Z/RAR

Images/scanned documents should use OCR when required.

Sensitive-Data Detection

Use appropriate combinations of:

- Regular expressions
- Checksum validation
- Named Entity Recognition
- Local AI/NLP if useful

Custom detection rules must be supported.

Example:

"CLIENT-12345678"

or

“CONFIDENTIAL ACQUISITION”

False positives should be minimized.

Exact File Location

When sensitive information is detected, the app should show:

- File name
- Full path
- File type
- Created/modified date
- Sensitive-data category
- Risk level
- Exact PDF page where possible
- Excel sheet/cell where possible
- Text line/paragraph where possible
- Masked evidence preview

Example:

Type: SIN
File: "C:\Clients\Client123\TaxReturn.pdf"
Page: 7
Preview: "*** *** 123"
Risk: High

Buttons:

- Open File
- Open Folder
- Mark Reviewed
- False Positive
- Ignore File
- Ignore Folder

Local Privacy

Default mode must keep files on the customer’s computer.

No document uploads to external AI/cloud services unless explicitly enabled in a future version.

Any external APIs/libraries must be disclosed.

The application should clearly show:

Local Scan – Documents remain on this computer

Performance

The app may need to process:

- 10,000 files
- 100,000 files
- 500,000 files
- 1,000,000+ files

It must not load everything into RAM.

Include performance modes:

Low Impact – computer remains usable
Normal – balanced
Overnight/Maximum – faster scanning

Allow control of:

- CPU usage
- Worker threads
- Disk activity where practical

Pause / Resume

Mandatory.

Scan progress must be saved continuously.

If a scan stops at 60% because of shutdown/restart/crash, reopening the application must allow:

Resume Previous Scan

It should continue rather than restart from zero.

Previously scanned unchanged files should not need complete rescanning.

Use file metadata/hash where appropriate.

Dashboard

Display:

- Total files discovered
- Files scanned
- Files remaining
- Completion %
- Current location
- Scan speed
- Findings count
- Access denied
- Encrypted/password protected
- Unsupported/corrupted files

Search & Filters

Filter findings by:

- Risk level
- Data type
- File type
- Folder
- Date
- File age
- Hidden/non-hidden
- Duplicate/non-duplicate
- Reviewed/not reviewed

Duplicate Detection

Identify exact duplicate sensitive files using file hashes.

Example:

“This passport scan exists in 6 different folders.”

Old Sensitive Files

Allow the user to flag sensitive files older than:

- 2 years
- 5 years
- 7 years
- 10 years
- Custom period

The software should only flag them for review, not claim they legally must be deleted.

Folder Permissions

Where technically possible, inspect Windows folder/file permissions.

Example:

Sensitive file located in folder accessible by 35 users

Flag as:

Potential broad-access exposure

Do not automatically classify this as a legal violation.

Quarantine

Allow users to move selected files to a quarantine folder.

Requirements:

- User confirmation required
- Record original location
- Record new location
- Audit log
- No automatic deletion

Review Actions

Allow:

- Reviewed
- Accepted
- False positive
- Needs action
- Ignored
- Notes/comments

Users should also be able to create ignore rules for specific folders/files/patterns.

Reporting

Export professional reports to:

- PDF
- Excel
- CSV

Reports should include:

- Total files scanned
- High/medium/low findings
- Duplicates
- Old sensitive files
- Permission exposures
- Files not successfully scanned
- Top risky folders
- Data categories detected

Full sensitive values should be masked.

Before/After Comparison

If the customer performs another scan after cleanup, show:

- Previous findings
- Current findings
- Removed findings
- New findings
- Remaining findings
- Percentage reduction

Audit Log

Record:

- Scan started/stopped/resumed
- Finding reviewed
- File quarantined
- Ignore rule created
- Report exported

Include date/time.

Local Database

Use SQLite or similar embedded database.

Store:

- File metadata
- Scan status
- Hash
- Findings
- Review status
- Scan history

Do NOT store unnecessary complete copies of customer documents.

Security

Requirements:

- No hard-coded passwords
- No hidden uploads/telemetry
- Mask sensitive previews
- Secure handling of temporary files
- Delete temporary extraction files when no longer required

Installer

Final product must include:

- Windows installer
- Uninstaller
- Application version
- App icon
- Installation documentation

Acceptance Testing

I will independently test the program before final payment.

I will create test folders with thousands of files containing fake sensitive information known only to me.

Tests may include:

- Fake SIN inside PDF
- Fake banking information inside DOCX
- Sensitive value inside XLSX
- Hidden file
- Deeply nested folder
- Image requiring OCR
- ZIP containing sensitive file
- Password-protected file
- Corrupted file

The developer will NOT know which files contain test information.

Readable/supported test files must be detected.

Unreadable files must appear clearly in the “not scanned” report.

Restart Test

I will:

1. Start a large scan
2. Stop/restart the computer before completion
3. Reopen the application

The program must resume the previous scan instead of starting again from zero.

Performance Test

Low Impact Mode will be tested while Chrome, Word, video and other normal applications are running.

I understand zero resource use is impossible, but the computer should remain reasonably usable.

Deliverables

I require:

- Finished Windows application
- Installer
- Full source code
- Build instructions
- Database schema
- Dependency/library list
- Technical documentation

I must have the legal right to commercially use, modify and sell the finished application.

All third-party libraries and licences must be disclosed.

Development Milestones

Suggested milestones:

1. Deep file crawler + file-status logging
2. PDF/Office/image content extraction
3. Sensitive-data detection
4. Findings UI + exact file location
5. Performance throttling + pause/resume
6. Duplicate/age/permissions analysis
7. Reports + audit log
8. Installer, testing, source-code delivery

Final payment only after acceptance testing.

Developer Requirements

Please apply if you have experience with:

- Windows desktop development
- Large filesystem scanning
- PDF/Office parsing
- OCR
- SQLite/local databases
- Multithreading
- Windows permissions
- Performance optimization
- Local NLP/AI
- Privacy/security software

IMPORTANT — Answer These Questions

Please include:

1. Programming language/framework you recommend
2. How you will scan 1,000,000 files efficiently
3. How you will prevent excessive slowdown
4. How pause/resume after restart will work
5. How hidden files will be scanned
6. How PDF/DOCX/XLSX content will be extracted
7. Which OCR engine you will use
8. How sensitive information will be detected
9. Whether any files ever leave the computer
10. How duplicates will be detected
11. How Windows permissions will be inspected
12. Which file types are difficult to support
13. Third-party libraries and their commercial licences
14. Confirmation that full source code will be provided
15. Fixed-price estimate and development timeline
16. Examples of similar work

Generic proposals that do not answer these questions may not be considered.

Please provide your proposed architecture, fixed price, timeline, milestones and similar previous work.