# Docs by LangChain: AGENT DEVELOPMENT LIFECYCLE > Documentation for LangSmith, Fleet, and our open source packages. ## AGENT DEVELOPMENT LIFECYCLE - [AGENT DEVELOPMENT LIFECYCLE / Build (451 pages)](https://docs.langchain.com/_llms/agent-development-lifecycle/build.txt): Build agents with open source - [AGENT DEVELOPMENT LIFECYCLE / Deploy (251 pages)](https://docs.langchain.com/_llms/agent-development-lifecycle/deploy.txt): Ship agents to production - [AGENT DEVELOPMENT LIFECYCLE / Monitor (668 pages)](https://docs.langchain.com/_llms/agent-development-lifecycle/monitor.txt): Trace, debug, and observe agents ### Home - [Docs by LangChain](https://docs.langchain.com/index.md) ### Test #### Get started - [LangSmith Evaluation](https://docs.langchain.com/langsmith/evaluation.md): Evaluate and test agent quality at scale with datasets, evaluators, prompts, and Studio. - [Evaluation quickstart](https://docs.langchain.com/langsmith/evaluation-quickstart.md) - [Evaluation concepts](https://docs.langchain.com/langsmith/evaluation-concepts.md) - [Application-specific evaluation approaches](https://docs.langchain.com/langsmith/evaluation-approaches.md) #### Datasets & Experiments ##### Datasets - [Manage datasets](https://docs.langchain.com/langsmith/manage-datasets.md) - [Custom output rendering](https://docs.langchain.com/langsmith/custom-output-rendering.md) ###### Create a dataset - [Create and manage datasets in the UI](https://docs.langchain.com/langsmith/manage-datasets-in-application.md) - [How to create and manage datasets programmatically](https://docs.langchain.com/langsmith/manage-datasets-programmatically.md) ##### Run an evaluation - [How to evaluate agents](https://docs.langchain.com/langsmith/evaluate-llm-application.md) - [How to evaluate with OpenTelemetry](https://docs.langchain.com/langsmith/evaluate-with-opentelemetry.md) - [How to use the REST API](https://docs.langchain.com/langsmith/run-evals-api-only.md) ##### Evaluation techniques ###### Define evaluation target - [How to define a target function to evaluate](https://docs.langchain.com/langsmith/define-target-function.md) - [How to evaluate an application's intermediate steps](https://docs.langchain.com/langsmith/evaluate-on-intermediate-steps.md) - [How to evaluate a runnable](https://docs.langchain.com/langsmith/langchain-runnable.md) - [How to evaluate a graph](https://docs.langchain.com/langsmith/evaluate-graph.md) - [How to simulate multi-turn interactions](https://docs.langchain.com/langsmith/multi-turn-simulation.md) - [How to evaluate your agent with trajectory evaluations](https://docs.langchain.com/langsmith/trajectory-evals.md) ###### Scoring methods - [How to return multiple scores in one evaluator](https://docs.langchain.com/langsmith/multiple-scores.md) - [How to return categorical vs numerical metrics](https://docs.langchain.com/langsmith/metric-type.md) ###### Experiment configuration - [Experiment configuration](https://docs.langchain.com/langsmith/experiment-configuration.md) - [How to run an evaluation asynchronously](https://docs.langchain.com/langsmith/evaluation-async.md) - [How to evaluate with repetitions](https://docs.langchain.com/langsmith/repetition.md) - [How to handle model rate limits](https://docs.langchain.com/langsmith/handle-model-rate-limiting.md) - [How to add evaluators to an existing experiment (Python only)](https://docs.langchain.com/langsmith/evaluate-existing-experiment.md) - [How to run an evaluation locally (Python only)](https://docs.langchain.com/langsmith/local.md) - [How to read experiment results locally](https://docs.langchain.com/langsmith/read-local-experiment-results.md) - [How to retry failed runs in experiments (Python only)](https://docs.langchain.com/langsmith/evaluate-with-retry.md) ###### Multimodal evaluations - [Run an evaluation with multimodal content](https://docs.langchain.com/langsmith/evaluate-with-attachments.md): Learn how to create dataset examples with file attachments and use them in prompts and evaluators when running LangSmith evaluations with multimodal content. ##### Analyze experiment results - [Analyze an experiment](https://docs.langchain.com/langsmith/analyze-an-experiment.md) - [How to compare experiment results](https://docs.langchain.com/langsmith/compare-experiment-results.md) - [How to filter experiments in the UI](https://docs.langchain.com/langsmith/filter-experiments-ui.md) - [How to fetch performance metrics for an experiment](https://docs.langchain.com/langsmith/fetch-perf-metrics-experiment.md) - [How to upload experiments run outside of LangSmith with the REST API](https://docs.langchain.com/langsmith/upload-existing-experiments.md) ##### Fine-tuning - [Fine-tune models with Smithtune](https://docs.langchain.com/langsmith/smithtune.md): Use Smithtune to fine-tune models on LangSmith conversations and compare the results in LangSmith. ##### Tutorials - [Evaluate a chatbot](https://docs.langchain.com/langsmith/evaluate-chatbot-tutorial.md) - [Evaluate a RAG application](https://docs.langchain.com/langsmith/evaluate-rag-tutorial.md) - [Test a ReAct agent with Pytest/Vitest and LangSmith](https://docs.langchain.com/langsmith/test-react-agent-pytest.md) - [Evaluate a complex agent](https://docs.langchain.com/langsmith/evaluate-complex-agent.md) - [Run backtests on a new version of an agent](https://docs.langchain.com/langsmith/run-backtests-new-agent.md) ##### Common data types - [Example data format](https://docs.langchain.com/langsmith/example-data-format.md) - [Dataset prebuilt JSON schema types](https://docs.langchain.com/langsmith/dataset-json-types.md) - [Dataset transformations](https://docs.langchain.com/langsmith/dataset-transformations.md) #### Evaluators - [Evaluation types](https://docs.langchain.com/langsmith/evaluation-types.md) - [Manage evaluators](https://docs.langchain.com/langsmith/evaluators.md): View and manage evaluators at the workspace level in LangSmith. - [Manage evaluators with the SDK](https://docs.langchain.com/langsmith/manage-evaluators-sdk.md): Create, retrieve, update, list, and delete LangSmith evaluators programmatically with the SDK. - [Automatically run evaluators on experiments](https://docs.langchain.com/langsmith/bind-evaluator-to-dataset.md) - [Track and limit evaluator spend](https://docs.langchain.com/langsmith/evaluator-spend.md): Cap weekly LLM spend on evaluators with an organization-wide default or per-evaluator overrides to keep evaluator costs predictable. ##### Evaluator types ###### UI - [How to define an LLM-as-a-judge evaluator](https://docs.langchain.com/langsmith/llm-as-judge.md) - [How to define a decision model evaluator](https://docs.langchain.com/langsmith/decision-model-evaluator.md): Set up a LangSmith evaluator that uses Jev as the judge, with noul, choice, and score questions that each return typed feedback. - [How to define a code evaluator](https://docs.langchain.com/langsmith/code-evaluator-ui.md) - [How to create a composite evaluator](https://docs.langchain.com/langsmith/composite-evaluators-ui.md) ###### SDK - [How to define an LLM-as-a-judge evaluator](https://docs.langchain.com/langsmith/llm-as-judge-sdk.md) - [How to define a code evaluator](https://docs.langchain.com/langsmith/code-evaluator-sdk.md) - [How to create a composite evaluator](https://docs.langchain.com/langsmith/composite-evaluators-sdk.md) - [How to define a summary evaluator](https://docs.langchain.com/langsmith/summary.md) - [How to run a pairwise evaluation](https://docs.langchain.com/langsmith/evaluate-pairwise.md) ##### Frameworks & integrations - [Run evals with openevals package](https://docs.langchain.com/langsmith/openevals.md): Run evaluations using the open-source openevals and agentevals packages with LangSmith. - [How to run evaluations with pytest](https://docs.langchain.com/langsmith/pytest.md) - [How to run evaluations with Vitest/Jest](https://docs.langchain.com/langsmith/vitest-jest.md) - [Harbor integrations](https://docs.langchain.com/langsmith/harbor-integrations.md): Run evaluations, Deep Agents, and sandboxes on LangSmith with Harbor. ##### Improve evaluators - [Improve LLM-as-judge evaluators using human feedback](https://docs.langchain.com/langsmith/improve-judge-evaluator-feedback.md) - [How to improve your evaluator with few-shot examples](https://docs.langchain.com/langsmith/create-few-shot-evaluators.md) - [How to audit evaluator scores](https://docs.langchain.com/langsmith/audit-evaluator-scores.md) #### Annotation Queues - [Use annotation queues](https://docs.langchain.com/langsmith/annotation-queues.md) - [Manage feedback & annotation queues programmatically](https://docs.langchain.com/langsmith/annotation-queues-sdk.md) - [Use assertions](https://docs.langchain.com/langsmith/assertions.md): Capture free-form acceptance criteria as dataset examples by writing assertions while reviewing runs in an annotation queue. - [Set up feedback criteria](https://docs.langchain.com/langsmith/set-up-feedback-criteria.md) - [Annotate traces and runs inline](https://docs.langchain.com/langsmith/annotate-traces-inline.md) ##### Feedback - [Log user feedback using the SDK](https://docs.langchain.com/langsmith/attach-user-feedback.md) - [Collect feedback with presigned URLs](https://docs.langchain.com/langsmith/presigned-feedback-tokens.md): Use presigned feedback tokens to collect user feedback from client-side applications without exposing your LangSmith API key. #### Test from Playground - [Test from the Playground](https://docs.langchain.com/langsmith/test-from-playground.md): Test prompts and model configurations over datasets in the LangSmith Playground without writing code. - [Run an evaluation from the Playground](https://docs.langchain.com/langsmith/run-evaluation-from-playground.md) #### Test from Studio - [LangSmith Studio](https://docs.langchain.com/langsmith/studio.md) - [Get started with Studio](https://docs.langchain.com/langsmith/quick-start-studio.md) - [How to use Studio](https://docs.langchain.com/langsmith/use-studio.md) - [Observability in Studio](https://docs.langchain.com/langsmith/observability-studio.md) - [Studio troubleshooting](https://docs.langchain.com/langsmith/troubleshooting-studio.md) ## OpenAPI Specs - [agent-server-openapi](/langsmith/agent-server-openapi.json) - [openapi](https://api.host.langchain.com/openapi.json) - [langsmith-platform-openapi](/langsmith/langsmith-platform-openapi.json)