langchain-course
A project-based course repository for developing AI agents using LangChain v1+ and LangGraph: search agents, RAG systems, reflection agents, and code interpreters.
- 1.8k
- Apache-2.0
- 2 d ago
A project-based course repository for developing AI agents using LangChain v1+ and LangGraph: search agents, RAG systems, reflection agents, and code interpreters.
🔍 AI observability skill for Claude Code. Debug LangChain/LangGraph agents by fetching execution traces from LangSmith Studio directly in your terminal.
Root cause analysis for AI agents. Detects agent loops, retry storms, and optimization opportunities in LangSmith, Langfuse, Arize Phoenix, and OpenTelemetry traces.
Workshop materials for Interrupt 2026: surfacing trends from agent traces with LangSmith Insights
A reference implementation for learning and building AI evaluation systems.
A hands-on session where you'll build a production AI agent from scratch using the Deep Agents framework, test and deploy it, and walk through LangSmith tracing and evaluations so you can observe and improve what you've shipped.
Shared code and tooling for building and signing LangSmith plugins
Best practices for evals from fundamentals to improving evals to production ci
Build and operate production-grade AI agent systems through this 22-chapter technical course designed for collaborative learning with AI partners.
Workshop materials for Interrupt 2026: aligning evals with LangSmith judges (PII detection and hand-off accuracy)
Self-directed Jupyter notebooks for engineers evaluating LangSmith during a POC. The modules cover the full agent engineering loop — build, trace, evaluate, deploy, and surface failure modes — against a single example agent.
Covering its core concepts, agent orchestration workflows, and essential tools. Design, implement, and manage multi-agent systems, integrate memory and tool-augmented reasoning, and deploy real-world AI agents efficiently. The course emphasizes hands-on, practical applications so you can immediately apply the knowledge to production-grade projects.
A 2-hour hands-on workshop: evaluation, monitoring, and observability for agents you actually ship. Python + LangSmith.
A parrot expert chatbot with intentional bugs, built to demonstrate LangSmith Engine's ability to identify issues in agent traces and propose fixes via PR.
AI credit analysis agent built with LangChain Deep Agents. Orchestrator coordinates research, data, and calculations subagents to produce formatted .docx memos with citation tracking and compliance guardrails. Full LangSmith tracing and an evaluator suite for quality testing.
requirements for AI initiatives against the EU AI Act, AICM, and AIEU frameworks. Built with LangChain, traced to LangSmith, compatible with LangGraph Studio. Supports Gemini, OpenAI, and Vertex AI.
Self-healing LangSmith eval pipeline which automatically diagnoses and fixes failing tool calls and improves your prompt as your AI Ecosystem grows
Always-ready LangSmith demo: a LangChain docs assistant built as a TypeScript Managed Deep Agent, with a React chat app
Demo of LangSmith alerting to Microsoft Teams channels using webhooks
Deep Agent that analyzes pharmaceutical patient support calls using three subagents: sentiment analysis), topic mining & adverse event detection, and agent performance review against SOP skills. Includes PII redaction and hallucination guardrails via middleware, audio file trace attachments, and a multi-prompt eval suite
Demonstrates the use of align evaluators in a step-by-step workflow
Agent which generates financial slides, multi-modal offline evals suite, and DataDog APM tracing with LangSmith agent tracing togetehr
Tracing IBM WatsonX Orchestrate Agents to LangSmith with OTEL
Run online evals outside LangSmith via webhook and attach scores back to traces — all API-controlled.