Deepchecks ORION: New SOTA for Evaluating Complex LLM Chains 👾

The upcoming Deepchecks event is about "Deepchecks ORION: New SOTA for Evaluating Complex LLM Chains".

In this webinar, we will talk about ORION, our new SOTA evaluation models designed for complex LLM chains, RAG pipelines, agentic workflows, and tool calling.

➡️ Register here: https://www.linkedin.com/events/deepchecksorion-newsotaforevalu7333064761611276289/
📆 Date & Time: June, 2024 | 5.00 PM CEST | 8.00 AM PST

Philip Tannor, CEO at Deepchecks, and Assaf Gerner, ML Researcher at Deepchecks, will walk through how ORION performs claim-level evaluations and how the system identifies failure modes such as hallucinations.

Topics that will be covered:

✅ How ORION evaluates outputs at the claim level with high precision
✅ How automated scoring logic identifies complex LLM behaviors like hallucinations, planning breakdowns, tool misuse, and more
✅ Deep dive into the Deepchecks "Swarm of Evaluation Agents" approach for evaluating LLM applications
✅ Integrations with popular stacks like LangChain, LlamaIndex, CrewAI, AWS AI Sagemaker, Autogen, and more.