How Pendo used LangSmith to trace Novus from user behavior to code fixes @LangChain
How Pendo used LangSmith to trace Novus from user behavior to code fixes  @LangChain
Uploaded July 2026 | Updated November 2026, 2 days ago
Zain Lakhani, Chief AI Officer at Pendo, shares how his team built Novus, a product agent that automatically instruments your product to surface usage insights and suggest fixes.

He walks through how LangSmith's trace dashboard became the core feedback loop for improving Novus: replacing manual design partner check-ins, driving eval set creation, and catching agent failures before customers encounter them. The result is a design partner base that has scaled directly alongside observability in the platform.

Watch his full testimonial below, and see more use cases at youtube.com/playlist?list=PLfaIDFEXuae2u29NANFbddyC0pkfGUKZ1

Chapters:
0:00 What Novus does
0:19 The hardest part of building an AI product
1:10 Why eval sets are the answer
1:32 How LangSmith replaced manual design partner check-ins
2:17 Catching failures before customers do
2:37 Why LangChain was the right framework for POC and production
2:50 How observability scaled the design partner base
3:05 LangSmith and Novus as complements
How Pendo used LangSmith to trace Novus from user behavior to code fixesDeploy Google ADK Agents on LangSmith Agent ServerGLM 5.2 + dcode: Frontier Coding with Open ModelsWhy Nobody Has Solved Continual LearningHow Do You Actually Evaluate an AI Agent?LangSmith Preview Builds: Test agent changes before productionHow Workday Builds Agent Prototypes in Under a Day with LangChainBuilding ToolLLM With LangGraph.jsManaged Deep Agents - IntroHow to Build Agents That Improve Over Time: Model, Harness, Context | Interrupt LondonHow To Build a Self-Improving Agent with LangSmith Engine and Context HubTool Problem or Model Problem? How Clay Tells the Difference
LangChain |

How Pendo used LangSmith to trace Novus from user behavior to code fixes

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER