Sentry spends $263 a day on Jev for 10M events
Sentry screens 10 million events a day through Jev at $263 a day to catch prompt injection, and is now testing its own classifier to hit the same accuracy for 90% less. The cheap model is cheap enough to meter, and still worth replacing.
Built with it
Treg ships open source people search on Jev at $0.0089 a lead
Treg released a Jev-powered people search that plugs into an agent, priced at $0.0089 per lead with the code open and no markup. He claims it is 85% cheaper than Clay and number one on people search bench accuracy. TypeSafe AI reposted it.
Just $0.0089/lead, No more $600 subscriptions,
Someone measured it
Sentry screens 10M events a day on Jev, now testing its own classifier
Sentry runs Jev over 10 million events a day to catch prompt injection before a coding agent reads an error report, and Cloudflare Developers says that costs $263 a day. The team is now testing its own classifier, aiming to match Jev's accuracy for 90% less. @grichadev wrote up how.
The next prompt injection your coding agent reads might be an error report.
Writing criteria
Theo turned his own agent rage moments into an 80-case tool eval
Theo rebuilt T3 Code's tool schemas against an eval he spent a week writing from the ways his agents used the tools plus every failure that annoyed him, and says tools are now used 4x better on 79% less context.
My eval had over 80 cases, and Opus failed over half of them.