Where chatbot data becomes clarity
Since 2019, we have been building tools and group learning environments that help teams read chatbot performance the way engineers read circuit diagrams — precisely, collaboratively, and with a clear sense of what to fix next.
See what we offerA team built around structured group work
Ualtimos Pirentex started as a small analytics consultancy focused on one specific problem: most teams measuring chatbot performance were doing it alone, without a shared framework or anyone to pressure-test their conclusions.
We built a group-first platform where practitioners work through real datasets together — 6 to 12 participants per cohort, 8-week structured programs, facilitated by specialists with at least 4 years of hands-on chatbot deployment experience. The format is deliberately tight because accountability and peer feedback are harder to fake in small groups.
- Countries represented in active cohorts 31
- Average cohort size 9
- Program weeks per cohort 8
- Facilitators with 4+ years experience 14
Chatbot analytics at scale
Group synergy is not a side effect — it's the mechanism
Most chatbot analytics training focuses on individual skill. We think that misses the point: real performance problems get solved when 3 people with different contexts look at the same data and disagree productively.
Each program runs on a fixed weekly rhythm — 2 live sessions per week, asynchronous dataset work in between, and a structured peer-review protocol on session 4 and session 7. Facilitators do not lecture; they surface disagreements and keep the group moving toward a shared diagnostic framework. Participants come from product, engineering, and operations backgrounds, and that mix is intentional — chatbot performance looks different depending on who's measuring it.
Program structure in detail
Darya Velychko
Lead Facilitator
Darya has run 28 cohorts since joining in 2020, with a background in NLP pipeline auditing. She designs the peer-review protocols used across all programs.
Ostap Hrynchuk
Analytics Curriculum Lead
Ostap built the diagnostic framework that participants use to evaluate chatbot drop-off rates, intent mismatch, and session depth across 11 metric categories.
Ivet Szabo
Group Process Specialist
Ivet focuses on the group dynamics side — how cohorts move from surface-level data reading to genuine diagnostic disagreement. She trains all incoming facilitators on the protocol.