All work
LLM & NLPResearch2026
Student PII detection & redaction
A FERPA-aligned model that finds and removes personal data from student writing before it reaches a third-party system.
Scope something similarCode walkthrough available on request
The problem
Education platforms want to use LLMs on student work, but sending names, emails and IDs to an external model is a compliance incident waiting to happen.
What we built
- Token-classification model fine-tuned on real annotated data (PIILO, CC BY 4.0) rather than synthetic examples
- Detection and redaction pipeline covering names, contact details, IDs and free-text identifiers
- Benchmarking against public leaderboard solutions so accuracy claims are comparable, not self-reported
- Designed around FERPA requirements, with the same pattern applying directly to GDPR workloads
Need something like this?
We'll tell you honestly whether your case is a two-week build or a three-month one — and what it would cost — before you commit to anything.
