Over the last few months building LLM features (support bots, summarizers, internal tools), we kept noticing the same pattern:

Applications were sending raw user data directly into LLM APIs things like names, emails, support tickets with account details, occasionally SSNs / phone numbers in logs

Not because people don’t care about privacy—mostly because it is pretty easy to pass JSON.stringify(data) into a prompt. Also, redaction is annoying and brittle (regex breaks fast)and there wasn’t a simple drop-in solution.

So we built a small middleware layer that sits between your app and the LLM. Give it a look and try out demo as well at https://getredacta.com