Hacker ApprovedDepartments

LangGraph

Ranked 4 of 28 agent frameworks by recommenders

Recommended for deterministic, stateful agent workflows with graph-based control, human-in-the-loop and time travel. Critics find it rigid and overly complex for simple tasks.

Recommended for deterministic, stateful agent workflows with graph-based control, human-in-the-loop and time travel. Critics find it rigid and overly complex for simple tasks.

54 people gave an opinion

recommend it
31
criticize it
23
approval share
57%
Each figure is one person. RecommendsCriticizes

Recommended in 12 of 13 months

  1. November 2025: recommended
  2. December 2025: recommended
  3. January 2026: recommended
  4. February 2026: recommended
  5. March 2026: recommended
  6. April 2026: recommended
  7. May 2026: recommended
  8. June 2026: recommended
  9. July 2026: recommended
  10. August 2026: recommended
  11. September 2026: recommended
  12. October 2026: recommended
  13. November 2026: not recommended (month in progress)

Why it doesn’t carry the badge

  • 31 people recommend itNeeds 10 or more, met.
  • 1.3 recommenders for every criticNeeds 2 to 1 or better, not met.
  • Recommended in 12 of 13 monthsNeeds 4 or more, met.

The same three rules apply to every product, and the badge can’t be bought. How the badge works

What people say about it

The reasons people give, with the number of comments that give each one. Open a reason to read what they said.

Why people recommend it

Each mark is one comment

  • Deterministic, controllable agent workflows

    “As long as you keep the flow rigid and only use the LLM for orchestration and formatting, the hallucinations pretty much disappear.”

    Comment from 23 Jan 2026 in “GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers”

    “I've been using LangGraph to enforce structured outputs and run validation checks before any side effects happen.”

    Comment from 26 Jan 2026 in “AI hallucinates. How do you keep it from fucking up automations?”

    “having that text state to pass into something like LangGraph is the only way to reliably control the logic”

    Comment from 24 Jan 2026 in “Ask HN: What's the current best local/open speech-to-speech setup?”
  • Graph and state machine model

    “We use LangGraph to implement the State Machine pattern rather than a linear Chain.”

    Comment from 9 Dec 2025 in “Show HN: Open-source reference architecture for AI Agents (LangGraph, Pydantic)”

    “the way to define the graph and transitions is pretty intuitive”

    Comment from 17 Dec 2025 in “Ask HN: What agent frameworks are you using, and how well do they work?”

    “it seems like the right abstraction to me”

    Comment from 12 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”
  • Reliable in production use

    “it seems to provide a lot of infrastructure for highly reliable and interactive agents”

    Comment from 11 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”

    “In our experience so far, it's been rock solid.”

    Comment from 16 Sep 2026 in “Show HN: Pizza Bot – An inbox for AI agents that work in the background”

    “I've not faced much issues!”

    Comment from 22 Nov 2025 in “Agent design is still hard”

What people warn about

Each mark is one comment

  • Adds complexity over plain Python

    “LangGraph wasting our time with what could've simply been an ordinary python function”

    Comment from 21 Feb 2026 in “Cord: Coordinating Trees of AI Agents”

    “LangGraph is unnecessary framework only slows you down by spiking up complexity.”

    Comment from 22 Jun 2026 in “Building Reliable Agentic AI Systems”

    “cobbled together mess using LangGraph and other overly complex libraries”

    Comment from 14 Sep 2026 on Hacker News
  • Rigid, less flexible design

    “it's more opinionated , less flexible”

    Comment from 12 Nov 2025 in “Adk-go: code-first Go toolkit for building, evaluating, and deploying AI agents”

    “they are inherently rigid”

    Comment from 21 Feb 2026 in “Cord: Coordinating Trees of AI Agents”

    “once the happy path breaks, the graph essentially halts or loops indefinitely because the error handling is too rigid”

    Comment from 12 Feb 2026 in “Show HN: Agent framework that generates its own topology and evolves at runtime”
  • Time lost on state management

    “I've spent way more time lately fighting with Celery queues and LangGraph state management”

    Comment from 26 Jan 2026 in “What "The Best" Looks Like”

    “Getting those agent systems to behave reliably was often annoying and tricky.”

    Comment from 11 Jul 2026 on Hacker News

Other agent frameworks people recommend

The four most recommended other agent frameworks, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.

LangGraph compared with 4 alternatives
ProductRecommendCriticizeApprovalMonthsOwnedBadge
Claude Agent SDK362559%11——
LangChain327929%101 yr3 owners—
Mastra32391%10—Approved
LangGraphThis page312357%12——
DSPy291960%8——
See all 28 agent frameworks, ranked

What people said

40 of the 63 comments that mention it, newest first. Each links to the original comment.

“frameworks like LangGraph etc provide a more sophisticated way of thinking about it”

RecommendsComment from 30 Oct 2026 in “Harness Engineering 101”

“In our experience so far, it's been rock solid.”

RecommendsComment from 16 Sep 2026 in “Show HN: Pizza Bot – An inbox for AI agents that work in the background”

“cobbled together mess using LangGraph and other overly complex libraries”

CriticizesComment from 14 Sep 2026 on Hacker News

“We already had the solution for compliance years ago with LangGraph”

RecommendsComment from 17 Aug 2026 in “Agent skills should be compiled, not just read”

“deterministic workflows such as what you describe is where langgraph really shines imo”

RecommendsComment from 15 Jul 2026 in “Towards a Harness That Can Do Anything”

“Getting those agent systems to behave reliably was often annoying and tricky.”

CriticizesComment from 11 Jul 2026 on Hacker News

“it felt too easy to accidentally miss them and end up reinventing the wheel”

MixedComment from 24 Jun 2026 in “Haystack: Open-Source AI Framework for Production Ready Agents, RAG”

“LangGraph is unnecessary framework only slows you down by spiking up complexity.”

CriticizesComment from 22 Jun 2026 in “Building Reliable Agentic AI Systems”

“For the agentic architecture we picked LangGraph — the modularity makes components easier to test.”

RecommendsComment from 13 Jun 2026 in “Show HN: Brightdeck – an OOXML-compatible AI presentation maker”

“port the agent logic into a framework like LangGraph or CrewAI”

MixedComment from 1 Jun 2026 in “Ask HN: What do you do while the coding agent is working?”

“has useful features like time travel, human in the loop capabilities (interruptions)”

MixedComment from 9 May 2026 in “Ask HN: What is the underlying stack behind multi-agent platforms?”

“langgraph is useful for deterministically reproducable orchestration”

RecommendsComment from 31 Mar 2026 in “The Claude Code Source Leak: fake tools, frustration regexes, undercover mode”

“they typically don't provide durable execution—if the process crashes, the agent will restart from scratch”

MixedComment from 12 Mar 2026 in “Running LangGraph, CrewAI, Google ADK with Durable Workflows”

“Leverage LangGraph's state machine to avoid OOM crashes.”

RecommendsComment from 5 Mar 2026 in “Show HN: L88 – A Local RAG System on 8GB VRAM (Need Architecture Feedback)”

“Visualize graphs using Graphviz – super visual and satisfying”

RecommendsComment from 24 Feb 2026 in “Agentic AI Tutorial: Step-by-Step Guide to Building Autonomous Agents (GitHub)”

“they are inherently rigid”

MixedComment from 21 Feb 2026 in “Cord: Coordinating Trees of AI Agents”

“LangGraph wasting our time with what could've simply been an ordinary python function”

CriticizesComment from 21 Feb 2026 in “Cord: Coordinating Trees of AI Agents”

“once the happy path breaks, the graph essentially halts or loops indefinitely because the error handling is too rigid”

CriticizesComment from 12 Feb 2026 in “Show HN: Agent framework that generates its own topology and evolves at runtime”

“I think the OpenAI Agents SDK is more model agnostic, LangGraph too.”

RecommendsComment from 6 Feb 2026 in “Ask HN: Best option for hosted agent in 2026?”

“I've been using LangGraph to force the model to extract and cite evidence before it runs any scoring logic.”

RecommendsComment from 26 Jan 2026 in “Notice of Collective Action Lawsuit Against Workday, INC”

“I've spent way more time lately fighting with Celery queues and LangGraph state management”

CriticizesComment from 26 Jan 2026 in “What "The Best" Looks Like”

“migrating back to LangGraph and Celery just to ensure reliability”

RecommendsComment from 26 Jan 2026 in “The Browser Is the Sandbox”

“I've been relying heavily on LangGraph and Celery to manage state”

RecommendsComment from 26 Jan 2026 in “The future of software engineering is SRE”

“I've been using LangGraph to enforce structured outputs and run validation checks before any side effects happen.”

RecommendsComment from 26 Jan 2026 in “AI hallucinates. How do you keep it from fucking up automations?”

“having that text state to pass into something like LangGraph is the only way to reliably control the logic”

RecommendsComment from 24 Jan 2026 in “Ask HN: What's the current best local/open speech-to-speech setup?”

“I switched from a similar parallelized setup to LangGraph precisely because the merge conflicts and redundant reasoning steps were killing my margins”

RecommendsComment from 23 Jan 2026 in “Multiclaude – Lightweight Multiagent Orchestrator”

“As long as you keep the flow rigid and only use the LLM for orchestration and formatting, the hallucinations pretty much disappear.”

RecommendsComment from 23 Jan 2026 in “GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers”

“found the general premise of the tools (in particular LangGraph) to be solid”

RecommendsComment from 26 Dec 2025 in “All I Want for Xmas Is Your Secrets: LangGrinch Hits LangChain (CVE-2025-68664)”

“I found the general premise of the tools (in particular LangGraph) to be solid.”

RecommendsComment from 26 Dec 2025 in “All I Want for Xmas Is Your Secrets: LangGrinch Hits LangChain (CVE-2025-68664)”

“the way to define the graph and transitions is pretty intuitive”

RecommendsComment from 17 Dec 2025 in “Ask HN: What agent frameworks are you using, and how well do they work?”

“The LangGraph docs do a pretty good job at convincing me that they got the latter pretty nailed down.”

MixedComment from 12 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”

“we didn't like the abstractions that many of the existing frameworks had built”

MixedComment from 12 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”

“it seems like the right abstraction to me”

MixedComment from 12 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”

“it seems to provide a lot of infrastructure for highly reliable and interactive agents”

RecommendsComment from 11 Dec 2025 in “Show HN: Sim – Apache-2.0 n8n alternative”

“We use LangGraph to implement the State Machine pattern rather than a linear Chain.”

RecommendsComment from 9 Dec 2025 in “Show HN: Open-source reference architecture for AI Agents (LangGraph, Pydantic)”

“Worked quite well, but the TS/JS langgraph version is way behind.”

MixedComment from 25 Nov 2025 in “Python is not a great language for data science”

“You will undoubtedly be recreating what already exists in LangGraph”

RecommendsComment from 23 Nov 2025 in “Agent design is still hard”

“I've not faced much issues!”

RecommendsComment from 22 Nov 2025 in “Agent design is still hard”

“just write the plumbing yourself, it's a lot easier than one might think”

CriticizesComment from 20 Nov 2025 in “Wrapping my head around AI wrappers”

“it's more opinionated , less flexible”

MixedComment from 12 Nov 2025 in “Adk-go: code-first Go toolkit for building, evaluating, and deploying AI agents”

Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.