The outcome

Compare a baseline and candidate with evidence strong enough to support a documented deployment decision.

LangSmith is a LLM application platform for tracing, observability, offline and online evaluation, prompt versioning, feedback, and agent deployment. Tracing agent and LLM runs, investigating failures, comparing prompts and models on datasets, collecting feedback, monitoring production quality, and promoting tested versions. This guide narrows that broad capability into one repeatable outcome, with checkpoints that keep the source material and your judgment in the loop.

Before you begin

Set the boundary before the tool starts.

Choose one real task, identify who will use the result, and decide what evidence or test will make the result acceptable. Gather only the source material needed for that task. If the work contains confidential, personal, regulated, or client-owned information, confirm that the platform and account are approved before sharing it.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

Step by step

A workflow you can repeat.

  1. 01

    Define the decision, user segments, critical tasks, prohibited failures, quality dimensions, latency and cost limits, evaluator owners, thresholds, and rollback rule.

  2. 02

    Build a versioned dataset from authorized synthetic, expert, and production-derived cases, removing secrets and personal data and preserving difficult counterexamples.

  3. 03

    Add deterministic code checks, structured human rubrics, and calibrated model judges, recording judge model, prompt, repetitions, known bias, and disagreement policy.

  4. 04

    Run baseline and candidate with identical data, concurrency, caching, and metadata, then inspect item-level regressions, confidence, variance, failure clusters, latency, and cost.

  5. 05

    Require owner sign-off on thresholds and exceptions, attach the experiment to the release record, canary the candidate, and feed verified production failures back into the dataset.

Working standard

What good use looks like.

  • Version data and evaluators together.
  • Inspect failures beyond averages.
  • Calibrate model judges against humans.

Tracing may capture prompts, responses, tool arguments, retrieved documents, personal data, secrets, or proprietary content. Retention differs by tier and datasets can outlive source traces. LLM-as-judge scores are model opinions and require calibration against human decisions. Use least-privilege keys and workspaces, conditional tracing, redaction, explicit retention, deletion tests, evaluator versioning, spend limits, and human review before treating a dashboard score as a release decision.

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this collection.