Blog
Highlights from the Currai blog: the posts worth reading first.
A complete, code-first guide to turning OpenAI Realtime voice conversations and text-to-speech generations into readable Currai User Stories and correctly nested traces.
A step-by-step guide to importing completed Retell AI calls into Currai with signed call-ended webhooks—without adding capture code to the live voice agent.
A step-by-step guide to importing completed Vapi calls into Currai with an authenticated webhook—without adding capture code or sharing a Vapi private API key.
Browse implementation notes, observability guides, product decisions, and workflow ideas by topic.
Design reusable LLM evaluators with stable rubrics, typed inputs, calibration examples, and versioning that work in development and production.
Read moreAn agent-generated result is not productive if a human must spend an hour repairing it. Add review, correction, rerun, and cleanup effort to your evaluation scorecard.
Read moreMulti-hour agents need evaluation at planning, execution, recovery, and handoff checkpoints. A final pass/fail score cannot explain where long-running work succeeds or collapses.
Read moreAgents that learn rules from feedback can improve quickly—and leak evaluation answers just as quickly. Separate learning, validation, and release data to measure generalization.
Read moreParallel agents can reduce wall time or multiply duplicated work, conflicts, and cost. Evaluate concurrency as a quality-adjusted scaling curve instead of assuming more agents are better.
Read more