AI BriefingLangChain Blog news
The latest AI and developer news from LangChain Blog, with short summaries.
Visit website →Breaking Through LangSmith's Limits with Replit Agent's Complex Workflows
Replit has upgraded LangSmith's capabilities to efficiently debug and monitor the complex workflows of Replit Agent.
Introducing Align Evals: Streamlining the LLM Application Evaluation Process
LangChain has launched Align Evals, a feature that lets you calibrate LLM evaluation scores to match human preferences.
What Is an AI Agent?
Rather than defining AI agents strictly, the concept of 'Agentic'—how autonomously a system operates—matters more.
Building a Trace Judge with Fireworks That Cuts Costs by 100x
LangSmith partnered with Fireworks to fine-tune a Qwen model, developing a 'Perceived Error' detection model that cuts costs by 100x.
Agentic Engineering: How Swarms of AI Agents Are Redefining Software Engineering
It proposes agentic engineering, which automates the entire software development lifecycle by using AI agents not as mere tools but as collaborating team members.
Agent Observability: How to Monitor and Evaluate LLM Agents in Production
Because LLM agents are non-deterministic and have an infinite input space, they require a different approach to observability than traditional software.
Aligning LLM-as-a-Judge with Human Preferences
LangSmith has introduced a feature that allows LLM evaluation models to improve themselves by learning from human feedback.
How to Debug and Evaluate AI Agents Using Observability: A LangChain Guide
Because AI agents follow a non-deterministic reasoning process, they require a different approach to observability and evaluation than traditional software.
How LangChain Made Coding Agent Spending Predictable
LangChain introduced the LangSmith LLM Gateway to implement real-time cost tracking and control features to prevent runaway cost increases from AI agents.
How Box AI Built Enterprise Content Agents with Deep Agents
Box built an agent system that performs complex enterprise data analysis by leveraging the Deep Agents architecture.
How to Choose the Right Sandbox for AI Agents
This explains the essential features and selection criteria for sandboxes to prevent security threats that arise when AI agents execute code.
How Benchling Builds Agents When Even the Best AI Isn't Smart Enough
Life sciences data platform Benchling is improving the accuracy of AI agents for scientific research through multi-model cross-validation and structured trace review.
The Missing Link Between Agents and Applications
LangChain overcomes the execution environment limitations of agents through Headless Tools, which can directly access client-side APIs and state.
Full-Text Search in SmithDB: Inverted Index Design for Object Storage
SmithDB designed an inverted index structure optimized for object storage to efficiently search agent traces with large JSON payloads.
Building an AI Health Coach: Evaluation, Safety, and Regulatory Compliance
Digital health company Numan explains how it built an automated evaluation loop using LLM personas to ensure the quality and regulatory compliance of its AI coach 'Nu'.
Give AI Agents Their Own Dedicated Computer
Introducing LangSmith Sandboxes, which provides hardware-level isolation so AI agents can safely and independently execute code.
LangGraph's Fault Tolerance: Retry, Timeout, Error Handler
LangGraph supports fault tolerance for real-world agents through Retry, Timeout, and Error Handler.
Model Neutrality: Why You Should Avoid AI Vendor Lock-in
As AI models become commoditized, model neutrality is becoming crucial to avoid being locked into vendors' orchestration tools.
How to Build a Custom Agent Harness
Covers how to build a custom agent harness using LangChain's create_agent and middleware.
Designing Efficient Verifiers for Legal Agents
Harvey and LangChain Labs studied how to design low-cost, high-reliability LLM verifiers for evaluating and post-training legal agents.