Cekura - Observe and analyze your voice and chat AI agents
by•
Out-of-the-box 30+ predefined metrics for analysis on CX, accuracy, conversation and voice quality. Compile perfect LLM judges by annotating just ~20 conversations and auto-improve in Cekura labs. Real-time, segmented dashboards to identify trends in Conversational AI. Smart statistical alerts so that you get notified only when metrics shift from historical baselines. Automated system pings to catch silent production failures.
Replies
Cekura
Cekura 🚀🚀🚀
A lot more to come!
Synthflow
Cekura is the best voice ai evals platform out there! We love it!
Cekura
Thanks@sassun - we love working with synthflow as well!
Cekura
Huge day for the Cekura team! 🚀 As a Founding AI Engineer here, launching this feels like a massive milestone.
The reality of Voice and Chat AI right now is that the tooling has lagged behind the models. We built Cekura because we knew that spot-checking a fraction of your calls just doesn't cut it anymore. Finding out your agent went rogue or dropped a compliance step days after the fact is brutal. Cekura exists to finally turn that chaotic wall of production data into clear, actionable truth.
On the technical side, I spent my time deep in the architecture of our metric feedback pipeline and structured simulations. These features completely eliminate "vibes-based" prompt tweaking by auto-compiling perfect LLM judges, while continuously mimicking production traffic to instantly catch silent infrastructure drops.
Really excited for you all to try it out. Let us know what you think of the new features! 🙌
So proud to be a part of the team! 🌟
Working in the frontend team, we have made sure all the heavy-duty production data actually feels intuitive to navigate, and the dashboard never feels overwhelming. We focused on building a UI that highlights those 'needle in a haystack' regressions—like missed disclosures—without the noise.
Can’t wait for everyone to explore the workspace!
Congrats on the launch! I would like to know the longest speech duration that can theoretically be analyzed. Could this be used to analyze speech recorded during a demo session with a customer? This product is able to analyze customer sentiment, which will be useful for sales.
this is cool. the 30 predefined metrics thing is smart cause most ppl building voice agents dont even know what to measure at first. nice that you dont have to start from scratch
We've been using Cekura for our voice AI testing and observability for the last year and the product is the best in the market. The team absolutely cooks!
Amazing product! Congrats team :) What are some of the voice only direct evals that the platform can perform?
Congratulations on your launch @kabra_sidhant. Trying to map this mentally—
Is Vocera closer to:
testing (like Playwright)
observability (like Datadog)
or eval frameworks (like DeepEval)?
Or is it a new category altogether?