Harness Hub
All startups

LangChain

The de-facto standard toolkit for building LLM applications and agents.

LangChain makes the LangChain framework and LangGraph, the low-level orchestration standard for stateful, controllable agents, plus the LangSmith platform for tracing, evaluating and monitoring them in production.

Pros & cons

Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.

Pros

  • DocsLangChain 1.0 (Oct 2025) is the first stable major release, with a commitment to no breaking changes until 2.0.[1][2]
  • DocsThe new create_agent API adds middleware hooks before and after model and tool calls, so the core loop can be customised.[1]
  • MetadataThe MIT-licensed repo is one of the most-used agent frameworks (about 149K GitHub stars), with commits landing daily.[1]
  • User reportsCompanion LangSmith tracing is praised as near-zero setup and stronger than cloud-native options such as AgentCore monitoring.[1][2]

Cons

  • User reportsOften criticised as a needless abstraction over vendor SDKs; many developers prefer calling provider packages directly.[1][2]
  • User reportsAbstractions get messy with complex state, according to production users.[1]
  • User reportsCompatibility with providers and proxies (OpenRouter, LiteLLM, OpenAI-compatible APIs) is a top open issue.[1]
  • User reportsUsers complain that docs and quickstarts increasingly assume LangSmith, its paid tracing product.[1]
Best for
Prototypes that lean on its integration catalogue, and teams already using LangGraph and LangSmith.
Watch out for
Abstraction overhead on stateful agents and gradual pull toward the paid LangSmith stack.

Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction

People

Official YouTube channel

Context engineering & memory · Evals & benchmarks

Viv

deepagents engineer

Harness engineering

Open roles (102)

All 102 roles

Their open-source work

Featured

langchain-ai/langgraph

Low-level orchestration for controllable, stateful agents.

Maintained by LangChain

1.2.24released
4 pros · 5 cons

Activity

Last push
· yesterday
Latest release
1.2.24
· yesterday
Last 90 days
20+ releases
  1. 1.2.24langgraph==1.2.24

    · yesterday

    Changes since 1.2.23 release(langgraph): 1.2.24 (#9752) fix(langgraph): handle REMOVE_ALL_MESSAGES in _messages_delta_reducer and None checkpoints in DeltaChannel (#9735)

  2. 1.2.23langgraph==1.2.23

    · 2d ago

    Changes since 1.2.22 release(langgraph): 1.2.23 (#9742) release(prebuilt): 1.1.3 (#9741) fix(langgraph,prebuilt): give Send-dispatched tool calls the same state as an edge (#9728) chore(deps): bump opentelemetry-api from 1.45.1 to 1.45.2 in /libs/langgraph (#9696)…

  3. prebuilt==1.1.3langgraph-prebuilt==1.1.3

    · 2d ago

    Changes since prebuilt==1.1.2 release(prebuilt): 1.1.3 (#9741) fix(langgraph,prebuilt): give Send-dispatched tool calls the same state as an edge (#9728) fix(prebuilt): make wrap_tool_call state overrides reach the tool (#9730)

  4. 1.2.22langgraph==1.2.22

    · 7d ago

    Changes since 1.2.21 release(langgraph): 1.2.22 (#9702) release(checkpoint-postgres): 3.2.1 (#9700) release(prebuilt): 1.1.2 (#9701) release(checkpoint): 4.3.2 (#9699) fix(langgraph): stream a child graph on its own thread to a caller streaming subgraphs (#9698)…

  5. checkpointpostgres==3.2.1langgraph-checkpoint-postgres==3.2.1

    · 7d ago

    Changes since checkpointpostgres==3.2.0 release(checkpoint-postgres): 3.2.1 (#9700) release(checkpoint): 4.3.2 (#9699) perf(checkpoint-postgres): run a lone store read outside a pipeline (#9635) perf(checkpoint-postgres): skip the async saver's lock for connection pools (#9638)…

Checked against GitHub on Nov 21, 2026. Release notes are shortened to plain text.

On X

Posts by the company and its team, shown as plain text; “Load on X” loads X's script for that post, which may set cookies.

  • 1️⃣ The new LangSmith Agent Platform

    2️⃣ New controls in LangSmith for protecting sensitive data in your traces

    3️⃣ LangSmith for Coding Agents

    4️⃣ LangSmith Tool Gateway

    💡 @jess_weng18 with a quick recap of everything we announced this week ⤵️ pic.twitter.com/cK9Cgvh5cO

    Open on X

  • to celebrate our fundraise we did a mini launch week this week (trace governance, tool gateway, LangSmith for coding agents

    to wrap it up, we're doing a webinar tmrw on the future of agents

    come join!https://t.co/emqlaieAec pic.twitter.com/qzw9sSSyLe

    Open on X

  • as agents get more goal oriented, having detailed control over the tools they use (and how they use them, and with whos auth they use them) is crucial https://t.co/kAifnRoXBC

    Open on X

  • great writeup! the alpha in harness engineering is in being a delivery mechanism for good context engineering

    the LLM context window is the sacred boundary where all computation happens -> the more we protect it and populate with the right stuff, the better our agents do https://t.co/p5R2JiHvSJ

    Open on X

  • New in LangSmith: Measure coding agent ROI

    📊 Our new dashboard pulls every coding agent you trace into one view, across harnesses + model providers

    ✅ Pull requests are now linked to threads, so you can see what a merged PR cost & how much spend never led to one… pic.twitter.com/npLhF8gYeo

    Open on X

  • How to get started

    1️⃣ Install the LangSmith plugin for each coding agent your team uses.

    Available for:
    ✅ Claude Code
    ✅ Codex
    ✅ Cursor
    ✅ GitHub Copilot
    ✅ Deep Agents

    2️⃣ Connect GitHub

    3️⃣ Open the coding agent overview page in LangSmith https://t.co/SoQfzexFs5

    Open on X

Follow @LangChain on X · The X feed

Videos

Their YouTube channel · All videos

In the news

LangChain Inc., a startup with a set of popular open-source tools for building artificial intelligence agents, has raised $300 million in funding. The Wall Street Journal reported today that Alphabet Inc.’s CapitalG fund led the round. It was joined by IVP, Sapphire Ventures, ...

In research (4)

Featured

Conversational AI agents for customer experience.

Sierra, co-founded by Bret Taylor and Clay Bavor, builds production conversational agents that resolve real customer issues with guardrails, evaluation and human escalation — a template for deploying vertical agents at the enterprise edge.

~$10B valuation (Oct 2025)

4 pros · 4 cons

Multi-agent orchestration, from open-source framework to enterprise platform.

CrewAI popularized role-based multi-agent teams ('crews') with a simple mental model and ships an enterprise control plane for deploying, monitoring and governing them.

$18M Series A led by Craft Ventures (2024)

3 pros · 4 cons

The data framework for connecting agents to your data.

LlamaIndex provides ingestion, indexing and retrieval primitives that give agents grounded context over enterprise data — a foundational piece of the retrieval side of the harness stack.

4 pros · 3 cons

Connections

Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.

The Harness Hub newsletter

New harness companies, projects, releases and the monthly State of Harness report, in your inbox.