HomeCompaniesfixa

sentry for AI voice agents

Fixa is the Sentry for AI voice agents. Companies like 11x and OfOne use us to run simulated tests, analyze production calls, and fix bugs in their voice agents. We were both voice agent developers at PlayHT – and, in order to find bugs in our agents, we would spend countless hours manually calling them and listening to conversation recordings. So, we’ve decided to build a platform that does this automatically. Fixa shows developers exactly where and how their agents mess up – enabling them to pinpoint root causes and fix them faster.
fixa
Founded:2024
Team Size:2
Location:San Francisco
Group Partner:Diana Hu

Active Founders

Oliver Wendell-Braly, Founder/CEO

helping voice agent developers sleep at night
Oliver Wendell-Braly
Oliver Wendell-Braly
fixa

Jonathan Liu, Founder/CTO

currently building fixa, the sentry for AI voice agents. helping voice agent companies find and fix bugs in their agents faster. previously the co-founder of schej, an open source scheduling website for college students and non-profits with over 30k monthly active users.
Jonathan Liu
Jonathan Liu
fixa

Company Launches

👋 hey everyone - we’re oliver and jonathan from fixa.dev

tl;dr

we find bugs in your voice agents - (including hallucinations, interruptions, transcription errors, and sentiment errors), so you don’t have to manually call your agents or listen to conversation recordings ever again. watch our launch video here


❌ the problem

finding bugs in voice agents is a giant hassle

traditionally, voice AI developers

  • hire teams of offshore 🌊 reviewers to read transcripts and listen in on calls
  • often don’t find bugs until a customer complains 😡
  • require the entire eng team to take 30 minutes out of their workday to call their agent prior to pushing changes 😪

✅ the solution

fixa finds bugs automatically in tests and production, providing developers with precise and actionable insights when things go south.

we pinpoint

  • LLM hallucinations 🤯
  • interruptions 🗣️
  • transcription errors 📝
  • sentiment errors (i.e. rude agent) 🤬

and propose changes to your prompt(s) accordingly.

we spent the past year developing voice agents at play.ai where we experienced the pain of manual bug hunting firsthand and we’re building the solution we wish we had. If you’re building in voice AI - we’d love to hear from you!

thanks ✌️