Hamming (@HammingAI) / X

Hamming

186 posts

Hamming

@HammingAI

We break AI voice agents before your customers do. Red-teaming • Pre-launch testing • Production monitoring

San Francisco

Joined May 2024

Hamming
@HammingAI
Jul 30
The real limit of manual voice-agent QA is what teams cannot test at all. @MavenAGI built its own simulations and smoke tests. Hamming turned them into release infrastructure for concurrency, noise, workflows, and regressions. “It’s not just time saving, it’s the capability and
Hamming
@HammingAI
Jul 28
Voice agents are built to help. That is also what makes them exploitable. Give an agent access to customer data and permission to act, and a normal conversation becomes an attack surface. Watch @sumanyu explain why teams should red-team voice agents:
Hamming
@HammingAI
Jul 23
A production voice agent needs more than a model. @OpenAI Presence shows what sits around it: policies, actions, evals, escalation rules, production signals, and controlled rollouts. Our addition: turn failed production calls into regression tests before the next release.
Hamming
@HammingAI
Jul 20
A voice agent can score 94% in the lab and 58% in a moving car. Only the background noise changed. Across 10M+ call minutes, ASR accuracy drops 20-40% once noise passes 10dB SNR. If you don't know your test calls' SNR, you don't know how the agent behaves in the real world. 👇
Hamming
@HammingAI
Jul 17
Compliance dashboards look great until the auditor asks a question they were never designed to answer. Which policy version? Who reviewed it? Was it fixed? Did it come back? We wrote a guide for building evidence packets that hold up. 👇
See @HammingAI's full profile
Sign up
Log in

See @HammingAI's full profile