AI: it's still just Engineering. If you disagree, if you think AI is completely new and old rules don't apply, use LangChain. If you think all the challenges of building robust applications still exist, and AI is just another (incredibly powerful) tool to manage, use Pydantic.
We’ve raised a $125M Series B at a $1.25B valuation to build the platform for agent engineering. Thank you to our investors (IVP, Sequoia Capital, Benchmark , Amplify Partners, Sapphire Ventures, CapitalG, and more) for your belief in us, and to our customers like Replit, Clay, Vanta, Cloudflare, Rippling, Cisco, Workday, and many more, for your trust and feedback along the way. Over the past few years, we’ve come to believe building agents requires a new discipline. We call it agent engineering, or the iterative process of refining non-deterministic LLM systems into reliable experiences. It combines aspects of product, engineering, and data science thinking, creating a way of working that produces more capable, more reliable agents. The space never stops, and neither do we. Since the launch of `langchain`, developers have told us they want more control, visibility, and flexibility when building agents. In rapid response, we built: • LangChain: Completely revamped in 1.0 with a core agent template and middleware for flexibility • LangGraph: A lower-level orchestration framework and runtime • LangSmith: Our platform for taking agents to production with observability, evaluation, and deployment This week, we’re launching new capabilities to help you iterate even faster: a new Insights Agent for aggregating usage patterns on production data, major 1.0 releases of LangGraph and LangChain, and our first no code text-to-agent builder, now in private preview. Read the full story and see what’s new: https://lnkd.in/gKVYYMUm