> ## Documentation Index > Fetch the complete documentation index at: https://launchdarkly.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Technical tutorials ### Explore end-to-end examples of how to build on the LaunchDarkly platform Dive into hands-on tutorials that demonstrate how to implement LaunchDarkly in real projects. Each tutorial is built to help you experiment quickly, adopt proven patterns, and ship with confidence. Learn how to forward Datadog application performance monitoring (APM) trace data to LaunchDarkly using the Datadog Agent, gate AI agent skills with feature flags, and set up dual shipping. OpenTelemetry is not required. Catch a dangerous agent skill before an agent runs it, using a security review from Tessl and a runtime check enforced by AgentControl. Optimize a reply-drafting agent to sound human, scored by GPTZero as an inverted judge, then run a fuller eval on the candidates it surfaces. Run four frameworks over one LaunchDarkly agent graph, let an experiment rank them on latency and tokens, then promote the winner with the same flag. Triage a real LaunchDarkly alert in Claude Code: correlate error groups, traces, and AgentControl variations to find the root cause. Set the foundation for an AI iteration loop with LaunchDarkly feature flags, AgentControl, and variation targeting, using a LangGraph decor agent. Walk the official langchain-ai/react-agent template through the migrate skill and externalize every prompt, model, and tool. Regression-test your support agent's RAG generation offline using Datasets, the Playground, and LLM-as-a-judge. Build multi-agent systems that work across LangGraph, OpenAI Agents SDK, and other frameworks. Learn how to instrument an AI agent to expose silent failures and use Vega AI to investigate the root cause. Build custom LLM-as-judge evaluators for AI code generation. Score security, API contracts, and scope creep. Instrument a real LLM application with OpenTelemetry spans and fan out traces to Langfuse and LaunchDarkly. Install LaunchDarkly Agent Skills in your AI coding assistant and create AgentControl configs from natural language. Learn how to use session replay and rage click detection to diagnose and fix production issues fast. Build a 3-agent research analysis swarm that runs across three orchestrators. Turn unstructured text into structured JSON with runtime-configurable schemas. Learn how combining session replay with online evaluations provides complete observability for multi-modal AI systems. Learn how to build SLOs that start with business impact and end with actionable targets. Santa's advice to engineering teams in need of unified observability solutions. Get feature flag context automatically injected into your OTel traces without writing instrumentation code. Learn 3 ways to stop cardinality from stealing your cloud budget. A full recap of the key innovations in observability this year. Why AI agents need three layers of observability when traditional observability can't help. Use LaunchDarkly's flags with the observability SDK to release and observe seasonal-themed CSS. Learn how users feel about your app's features so you can make data-driven product decisions. A practical guide for LLM observability and real-time quality monitoring in production. Learn how to automatically detect user frustration with LaunchDarkly's session replay and rage clicks. Connect rage click detection to LaunchDarkly Guarded Releases for automating actions. See how your React Native iOS app is performing with LaunchDarkly's Observability SDK. Build a CI/CD pipeline that validates and tests LaunchDarkly AgentControl configs before deployment. Learn about how LaunchDarkly's foundations works under the hood. Transform your basic multi-agent system with business tiers and geographic targeting. Build a multi-agent system with dynamic configuration using LangGraph and LaunchDarkly AgentControl configs. Walk through an integration between Snowflake Cortex and LaunchDarkly's AI SDKs. Learn how to use LaunchDarkly AgentControl configs to review and validate database schema changes. Learn to use a websocket library to update your Python Flask app when a kill switch flag is flipped. Tips and tricks to make Cursor more accurate, personalized, and extensible. Create, evaluate, and modify flags from within your IDE using natural language. GORM is a popular ORM for Go that allows you to interact with databases using Go structs. OpenTelemetry defines how we send telemetry data to observability backends. Collect traces, metrics, and logs from your React Native applications. Learn how to send telemetry data from Python applications. Leverage OpenTelemetry on the client side to collect signals from your applications. Observability is key to monitoring and improving your web application's performance. OpenTelemetry origins, collector setup, and how to start collecting OTEL data in your applications. Learn how to implement distributed tracing in your Next.js applications. Real-time monitoring for maintaining performance and reliability of Django applications. Compare the performance of DeepSeek r1 and Alibaba's Qwen using Ollama and AgentControl configs. Implement application tracing in .NET for performance monitoring. In any Ruby application, logging is more than just a means to record errors. Learn how to use materialized views in ClickHouse for better performance. Advanced ingestion filtering techniques for LaunchDarkly. Looking to set up your own Kafka cluster on AWS MSK? Trying to publish an npm package but have a complicated monorepo setup? Chrome Developer Tools is a toolkit for web developers built into Google Chrome. Data validation is critical for any application that relies on input from users. Use JavaScript or TypeScript for both your frontend and your backend. At LaunchDarkly, we're focused on keeping our app snappy and fast. A holiday-themed video illustrating the value of connecting feature flags with observability. Learn how users feel about your app's features so you can make data-driven product decisions. Safely and efficiently run a stage migration and monitor metrics. Use LaunchDarkly to create segments that target different users in production. Learn about how LaunchDarkly's foundations works under the hood. How to vibe code safely using LaunchDarkly feature flags. Create a kill switch that disables a feature in production. Use LaunchDarkly to roll out a feature safely. Learn to use a websocket library to update your Python Flask app when a kill switch flag is flipped. Create, evaluate, and modify flags from within your IDE using natural language. Catch a dangerous agent skill before an agent runs it, using a security review from Tessl and a runtime check enforced by AgentControl. Optimize a reply-drafting agent to sound human, scored by GPTZero as an inverted judge, then run a fuller eval on the candidates it surfaces. Run four frameworks over one LaunchDarkly agent graph, let an experiment rank them on latency and tokens, then promote the winner with the same flag. Walk the official langchain-ai/react-agent template through the migrate skill and externalize every prompt, model, and tool. Regression-test your support agent's RAG generation offline using Datasets, the Playground, and LLM-as-a-judge. Build multi-agent systems that work across LangGraph, OpenAI Agents SDK, and other frameworks. Skip the UI. Tell your AI coding assistant what to build and watch it create your agents. Build custom LLM-as-judge evaluators for AI code generation. Score security, API contracts, and scope creep. Learn how to use LaunchDarkly Judges to automatically evaluate AI model responses. Install LaunchDarkly Agent Skills in your AI coding assistant and create AgentControl configs from natural language. Build a 3-agent research analysis swarm that runs across three orchestrators. Turn unstructured text into structured JSON with runtime-configurable schemas. Explore the three 'ghosts' haunting AI in production: model, knowledge, and user layer. Learn how combining session replay with online evaluations provides complete observability for multi-modal AI systems. Why AI agents need three layers of observability when traditional observability can't help. A practical guide for LLM observability and real-time quality monitoring in production. Build a CI/CD pipeline that validates and tests LaunchDarkly AgentControl configs before deployment. Transform your basic multi-agent system with business tiers and geographic targeting. Build a multi-agent system with dynamic configuration using LangGraph and LaunchDarkly AgentControl configs. Walk through an integration between Snowflake Cortex and LaunchDarkly's AI SDKs. Learn how to use LaunchDarkly AgentControl Configs to review and validate database schema changes. Compare the performance of DeepSeek r1 and Alibaba's Qwen using Ollama and AgentControl configs. Learn how to instrument an AI agent to expose silent failures and use Vega AI to investigate the root cause. Learn how combining session replay with online evaluations provides complete observability for multi-modal AI systems. Instrument a real LLM application with OpenTelemetry spans and fan out traces to Langfuse and LaunchDarkly. Implement application tracing in .NET for performance monitoring. Learn how to implement distributed tracing in your Next.js applications. Real-time monitoring for maintaining performance and reliability of Django applications. Advanced ingestion filtering techniques for LaunchDarkly. Collect traces, metrics, and logs from your React Native applications. Observability is key to monitoring and improving your web application's performance. Leverage OpenTelemetry on the client side to collect signals from your applications. GORM is a popular ORM for Go that allows you to interact with databases using Go structs. In any Ruby application, logging is more than just a means to record errors. OpenTelemetry defines how we send telemetry data to observability backends. Learn how to send telemetry data from Python applications. Learn how to trace distributed systems in your Next.js applications. OpenTelemetry origins, collector setup, and how to start collecting OTEL data in your applications. Learn how to use session replay and rage click detection to diagnose and fix production issues fast. A holiday-themed video illustrating the value of connecting feature flags with observability. Learn how to build SLOs that start with business impact and end with actionable targets. In this series, we've talked about how observability has changed in 2025. Learn to apply a popular tidying method to consolidate an inflated observability stack. Santa's advice to engineering teams in need of unified observability solutions. Get feature flag context automatically injected into your OTel traces without writing instrumentation code. Learn 3 ways to stop cardinality from stealing your cloud budget. A full recap of the key innovations in observability this year. Why AI agents need three layers of observability when traditional observability can't help. Use LaunchDarkly's flags with the observability SDK to release and observe seasonal-themed CSS. Learn how to automatically detect user frustration with LaunchDarkly's session replay and rage clicks. Connect rage click detection to LaunchDarkly Guarded Releases for automating actions. See how your React Native iOS app is performing with LaunchDarkly's Observability SDK. Tips and tricks to make Cursor more accurate, personalized, and extensible. Chrome Developer Tools is a toolkit for web developers built into Google Chrome. Learn how to use materialized views in ClickHouse for better performance. Looking to set up your own Kafka cluster on AWS MSK? Trying to publish an npm package but have a complicated monorepo setup? Data validation is critical for any application that relies on input from users. Use JavaScript or TypeScript for both your frontend and your backend. At LaunchDarkly, we're focused on keeping our app snappy and fast. Skip the UI. Tell your AI coding assistant what to build and watch it create your agents. Learn how to use LaunchDarkly Judges to automatically evaluate AI model responses. A holiday-themed video illustrating the value of connecting feature flags with observability. Explore the three 'ghosts' haunting AI in production: model, knowledge, and user layer. In this series, we've talked about how observability has changed in 2025. Learn to apply a popular tidying method to consolidate an inflated observability stack. Safely and efficiently run a stage migration and monitor metrics. Use LaunchDarkly to create segments that target different users in production. How to vibe code safely using LaunchDarkly feature flags. Create a kill switch that disables a feature in production. Use LaunchDarkly to roll out a feature safely.