AI Agents in Production: Lessons from Rippling and LangChain

How does a company deploy AI agents across HR, payroll, IT, and finance products used by thousands of companies? Ankur Bhatt, Head of AI at Rippling, shares insights on building production-ready agents with Harrison Chase, CEO of LangChain.

In this conversation, we'll walk through:

- Evolving from simple AI features to complex agents—and why "deep agent" paradigms that leverage LLM reasoning outperform rigid workflow-based approaches

- Getting agents to production: using LangSmith to trace production behavior and debug agent execution, combined with real production data and rapid user feedback loops

- Scaling AI development: managing 150+ hack week projects while balancing bottom-up innovation with strategic product priorities

🔗 Resources mentioned:

LangChain: https://www.langchain.com
LangSmith: https://smith.langchain.com
LangGraph: https://langchain-ai.github.io/langgraph/
Rippling: https://www.rippling.com