Webinar: Automate evals & observability with Claude Code + orq.ai
About this event
In this hands-on session, we'll show you how to connect Claude Code directly to your orq.ai platform. Use it to build, run, and analyze evaluations and talk to your observability data - without leaving your terminal.
This isn't a slideshow. We'll walk through real workflows, live, and take your questions at the end.
What you'll learn
Connect Claude Code to your orq.ai platform
Set up the connection between Claude Code and orq.ai so you can query your production data, run experiments, and analyze results directly from your terminal.
Build and run evaluations through MCP
Use Claude Code to create and execute evaluations on orq.ai. Set up Python evaluators, LLM-as-a-Judge scoring, and JSON validators - then run them against your data.
Talk to your observability data
Go beyond dashboards. Ask Claude Code to analyze your traces, identify failure patterns, compare model performance across configurations, and surface issues you'd otherwise catch manually - or miss entirely.
Close the loop between observability and action
See how production insights feed directly into your evaluation workflow. Spot a regression in your traces, build an experiment to validate it, and iterate - all in one session, all from the CLI.
Who should attend
ML Engineers and Developers building and maintaining LLM-powered applications
Product Teams shipping AI features who need confidence in output quality
Engineering Leaders establishing evaluation workflows and quality standards for AI systems
Anyone sitting on orq.ai observability data and wanting to do more with it than stare at dashboards
Format
Live demo and hands-on walkthrough (approximately 45 minutes), followed by a Q&A session (approximately 15 minutes).