MaskLayer Studio is a local-first redaction workspace for preparing sensitive datasets before AI model training.
What it does: It lets users upload TXT, CSV, or JSONL files, detects sensitive information like emails, phone numbers, API keys, cards, URLs, IPs, and internal project terms, then helps review and redact those findings before export.
Why I made it: AI teams often work with messy training data that may contain private or risky information. I wanted to create a lightweight tool that makes dataset cleanup more visible, reviewable, and safer before that data enters a training pipeline.
How it works: The workspace runs in the browser. Users upload a dataset, review detected findings, choose redaction modes like masking, removal, partial masking, or stable tokens, then generate a cleaned output and privacy report. The interface also includes an evaluation tab to check dataset readiness, unresolved risks, and export status.
MaskLayer Studio is a local-first redaction workspace for preparing sensitive datasets before AI model training.
What it does: It lets users upload TXT, CSV, or JSONL files, detects sensitive information like emails, phone numbers, API keys, cards, URLs, IPs, and internal project terms, then helps review and redact those findings before export.
Why I made it: AI teams often work with messy training data that may contain private or risky information. I wanted to create a lightweight tool that makes dataset cleanup more visible, reviewable, and safer before that data enters a training pipeline.
How it works: The workspace runs in the browser. Users upload a dataset, review detected findings, choose redaction modes like masking, removal, partial masking, or stable tokens, then generate a cleaned output and privacy report. The interface also includes an evaluation tab to check dataset readiness, unresolved risks, and export status.