Top AI Product

Teams shipping LLM features usually bolt together three or four separate tools: one for logging, another for evaluations, a third to route between providers. Respan, a Y Combinator-backed platform formerly known as Keywords AI, folds all of that into a single LLM engineering layer for observability, evals, prompt optimization, and an AI gateway.

## What Respan does

You send OpenAI-style calls through Respan and it routes them to 500+ models, logging every request along the way. Dashboards slice usage by model or user, showing requests, tokens, errors, latency, and cost in one place. Evaluation workflows combine human review, code checks, and LLM judges against the metrics you care about. The company says it now processes over 1 billion logs and 2 trillion tokens a month across 100+ AI startups and enterprise teams.

## The gateway and API

Respan’s gateway is the entry point: point your existing OpenAI SDK at it, or keep each provider’s native SDK on a passthrough endpoint. Either way every call is captured for observability and evals, so adding the layer does not mean rewriting your stack. Respan raised a $5M seed round in March 2026.