TrustRadius: an HG Insights company

LangSmith

Score9 out of 10

16 Reviews and Ratings

What is LangSmith?

LangSmith, by LangChain is a framework-agnostic Agent Engineering Platform designed for the observation, evaluation, and deployment of Large Language Model (LLM) agents. The solution provides a unified environment to transform Trace Data into actionable insights for iterative agent improvement and enterprise-scale production management.

Read more details.

Categories & Use Cases

Who Buys & Uses LangSmith Agent Engineering Platform

LangSmith Reviews

What users are saying about LangSmith.
View all LangSmith reviews

Game Changer for Agentic Workflows and Quality Control

Purna Devi Kiran Kosanam
Software Engineer in Engineering at Edtex (11-50 employees)

Use Cases and Deployment Scope

We use LangSmith Agent Engineering Platform for build an AI answer engine for our product which is like a customized AI chatbot for respective products which can have respective data of clients. LangSmith Agent Engineering Platform helped us to solve the problem in a better way. The capabilities of the LangSmith Agent Engineering Platform enabled us to make the AI answer engine.

Pros

  • LangChain and LangGraphe Integration
  • Production Monitoring
  • Evaluation and Testing
  • End to End tracing

Cons

  • When AI agents take multiple steps or use several tools the trace view becomes hard
  • It is harder to use other frameworks that are outside of LangChain. It requires lot of extra setup
  • Creating test datasets and custom rules feels hard
  • Tracking of usage and cost is little confusing, especially for agents that execute small intermediate steps

Return on Investment

  • It almost reduced 50% of our debugging time and helped us to speed up our feature delivery
  • It lowered our LLM costs
  • It has faster and safer updates

Usability

Other Software Used

Anthropic Claude, Cursor, ChatGPT

LangSmith Agent Engineering Platform From Pioneers to Undisputed Market Leaders

Incentivized
Rajveer MathurView profile
Senior AI/ML Engineer in Engineering at Brillio (5001-10,000 employees)

Use Cases and Deployment Scope

I was a user of LangChain from pioneer years, but LangSmith Agent Engineering Platform has being the game-changing addition into the picture. With LangSmith Agent Engineering Platform I just add one line into the codebase and then prompt management and API tracing happens. In my organization Prompt management and API tracing has been constructively the faster go-to-market solution, as we change the prompts from Prompt Hub and API tracing when we want to debug or change LLMs helps on our template codebase to deploy faster for client displays.

Pros

  • Prompt Hub usage when we only need to change prompts on deployed solutions. This is best on faster debugging cutting time by 50%.
  • API tracing dashboard provides instant telemetry of the API segment results and other metrics which helps us to strategize multiple approaches to cost-saving, time complexities, etc.
  • The Agent Builder which delivers all wiring and agents with only NLP descriptions. No custom adapters, no prompt management, no tools development.

Cons

  • UI/Dashboards used to change quite frequently, which leads into learning curves every week.
  • MCP servers were not visible when I used to reach the end of Agent Builders.

Return on Investment

  • The debugging time in terms of production failures for any AI Systems has been reduced approximately by 60%.
  • Agent Builder constructs full AI systems which are relatively simple and basic (Email triage/consolidator) in almost no time, quantitively reducing the development time by 80%

Usability

Alternatives Considered

Langfuse

Other Software Used

Ollama, Langfuse

Langfuse All rounder tracker to agentic applications.

Incentivized
Verified User
Employee in Engineering (51-200 employees)

Use Cases and Deployment Scope

LangSmith is very helpful in our organization for observability and tracking agent performance. Our scope uses agent development life cycle from expermentation to real production. We can find bugs and model usage costs, and the application is easy to track. It's the best place to improve performance and track the application.

Pros

  • Provide detailed end to end traces of our agentic application.
  • Systematic procedure to evaluation of the agentic and AI applications.

Cons

  • LangSmith can be improved with more interactive UI.
  • Documentation of the LangSmith can be include with more advance agentic insights.

Return on Investment

  • LangSmith reduces the time we spend troubleshooting agent behavior and helps the team debug faster.
  • It helps in trace, debug and evaluate agentic workflows.

Usability

Alternatives Considered

Langfuse

Other Software Used

Langfuse

Fast and Furious to Observability in AI application.

Incentivized
Verified User
Engineer in Engineering (51-200 employees)

Use Cases and Deployment Scope

LangSmith is the best tool for capturing analytics of my AI projects. Let me tell you the problem that is solved: how my LLM is used, how much input it can take, and how much output it can generate. The best thing about LangSmith is how easy it is to integrate. LangSmith perfectly traces my application along with costs.

Pros

  • Debugging make simple by using Lang Smith.
  • Easy to integrate.
  • Cost Effective to use.

Cons

  • Mention different types of eval factors.
  • Create more dynamic dashboards based on data.

Return on Investment

  • Simple and Easy.

Usability

Alternatives Considered

Langfuse

Other Software Used

Langfuse, PostHog