Skip to content

Notes from the team.

What we're learning about AI agents, messaging, voice and building software that lasts.

  1. AI & MLDesigning autonomous voice agents that feel humanWhat it actually takes to build voice agents that handle real calls end to end — latency budgets, turn-taking, and graceful failure.Nino Beridze8 min read
  2. EngineeringShipping with LLMs in production without losing your mindEvaluation, guardrails and observability are what separate an LLM demo from a product you can actually operate.Luka Tabidze7 min read
  3. CompanyWhy we build end to endOne team owning product, design, AI and infrastructure ships faster and breaks less. Here is how we structure it.Mariam Kapanadze5 min read
  4. ProductMeasuring the ROI of voice automationA practical framework for deciding which calls to automate first — and how to prove the impact.Giorgi Mamaladze6 min read
  5. DesignA design system that scales with the productHow we keep a fast-moving product visually coherent with primitives, tokens and ruthless consistency.Ana Dvali5 min read
  6. EngineeringEngineering for reliability from day oneReliability is a design constraint, not a phase. The practices we bake in before the first user arrives.Luka Tabidze6 min read