LangChain

AIAI-generated summary LangChain says its GTM Agent has saved sales representatives thousands of hours over the last 12 weeks while generating 1,540 sales artifacts. LangChain explains how business context and the Deep Agents harness were used to build artifact generation for sales and marketing workflows.

Agent Frameworks Workflow Automation

AIAI-generated summary LangChain discusses techniques for agent engineering to manage model context windows and reduce context rot during longer tasks. The open source harness Deep Agents uses middleware to move context out of active windows into files, summaries, or separate agents.

Agent Frameworks Agent Memory Workflow Automation Context Engineering

AIAI-generated summary Workday built an enterprise-scale AI agent platform utilizing LangGraph for orchestration and LangSmith for tracing, observability, and evaluations. Workday says the new platform reduces prototype development time from months to under a day while satisfying strict security requirements.

Agent Frameworks Agent Orchestration Enterprise AI

AIAI-generated summary Stripe built its Knowledge AI Platform, called Kai, in one week using LangChain's Deep Agents agent harness. Stripe says adoption of Kai grew from 296 users to over 5,000 in roughly four weeks after its release.

Agent Frameworks Workflow Automation AI Adoption

AIAI-generated summary Dosu says that building production-grade language model products is difficult because ensuring reliability is far from straightforward. Dosu says that LangSmith has allowed the company to scale evaluation driven development by enabling easy monitoring and searching of activity.

Benchmarks

AIAI-generated summary LangChain has released Promptim, an experimental open-source library designed to help developers systematically improve AI systems through automated prompt optimization. The library integrates with LangSmith to support dataset management, tracking results, and incorporating human feedback during the optimization loop.

APIs & SDKs Prompt Engineering

AIAI-generated summary LangChain explains eval-driven development as a process using a consistent set of metrics to measure AI agent performance and guide decision making. LangChain says the approach borrows from test-driven development and breaks down into defining agent behavior, building a dataset, and writing an evaluator.

Agent Frameworks Datasets Agent Evaluation

AIAI-generated summary Langchain says standard ReAct loops block long tool calls and queue other tasks, so the ecosystem is converging on a split between user-facing agents and background workers. Langchain says it is bringing this pattern to Deep Agents with async subagents, where an orchestrator stays available to the user while background workers run tasks in parallel.

Agent Frameworks Multi-Agent Systems Agent Orchestration

AIAI-generated summary LangChain discusses how agents navigate limited context windows by using retrieval techniques and agent harnesses. LangChain explains that tools like semantic search, filesystem indexing, and progressive disclosure help agents find necessary information.

Agent Frameworks Agent Memory RAG Context Engineering

AIAI-generated summary LangChain discusses how human review helps teams define quality and evaluate agent performance in context. Fetch Rewards uses annotation queues in LangSmith to turn human judgment into a reference point for automated judges.

Workflow Automation Agent Evaluation

AIAI-generated summary LangChain has made LangSmith Bring Your Own Cloud available to run on AWS, Azure, and Google Cloud. LangChain says the offering keeps sensitive data in customer cloud accounts while the company manages the data plane for orchestration and scaling.

APIs & SDKs

AIAI-generated summary Langchain launched Managed Deep Agents last week, providing users with a new way to give their agents a browser. Langchain says the tools behind the browser help agents manage context efficiently and complete multi-step work.

Agent Frameworks Browser Agents

AIAI-generated summary LangChain previewed Managed Deep Agents in public beta on LangSmith Cloud to let developers build and deploy agents with a folder of files and one command. LangChain says Managed Deep Agents runs on LangSmith Agent Server to provide durable threads and streaming runs without operating a server.

Agent Frameworks Deep Research

AIAI-generated summary LangChain launched browser use capabilities for Managed Deep Agents to enable developers to build web-enabled agents. LangChain says the setup supports tasks like booking reservations and end-to-end testing while managing both the browser and its toolset.

Agent Frameworks Browser Agents