LangChain

What hackers pick instead

• Claude Agent SDKDevelopers praise it for building agents on top of Claude Code at a predictable cost. Critics agree the main caveat is Anthropic's shifting rules on using subscriptions with it.36recommend
• MastraOpen-source TypeScript framework for building AI agents, praised for workflow primitives and developer experience. Common caveat: its opinionated patterns can slow you down when your project doesn't fit them.32recommend
• LangGraphRecommended 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.31recommend
• DSPyPeople use it to structure LLM calls as typed programs and optimize prompts. Critics say its value depends on a good automated metric and up-front setup.29recommend
• Pydantic AIDevelopers like it for handling the tool-call loop, structured outputs and typed LLM calls with little to learn. Caveat: good observability requires the paid SaaS.27recommend

Every quote

The 40 most detailed of 126 comments. Each links to the original.

1. I could maybe see that being the case for LangChain, specifically, if you're just thinking about abstracting the LLM provider's APIs.

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

2. The prototype worked great: chain a few calls together, get a response and it felt like we were basically done.

   Mixed · Mar 2026 · in “LangChain is powerful, but running it in production isn't”

3. The company I work for uses LangChain heavily, but that's because we have fairly complex requirements

   Recommends · Feb 2026 · in “Does anyone use CrewAI or LangChain anymore?”

4. for all the crap that LangChain gets it at least has its own layer for upstream LLM provider calls

   Mixed · Mar 2026 · in “LiteLLM Python package compromised by supply-chain attack”

5. LangChain is definitely a bloated mess, but it provides structure

   Mixed · Feb 2026 · in “Your Agent Framework Is Just a Bad Clone of Elixir”

6. Langchain is a decent place to start if starting from scratch.

   Recommends · Dec 2025 · in “Ask HN: What framework/tool should I use for building agents?”

7. It’s deployed and doing what everyone needs it to do.

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

8. they do have good prompts backed into the framework

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

9. I appreciate having a unified API across providers

   Mixed · Jun 2026 · in “Haystack: Open-Source AI Framework for Production Ready Agents, RAG”

10. which already has most of the facilities for that

    Recommends · Nov 2025 · in “Launch HN: Onyx (YC W24) – Open-source chat UI”

11. langchain makes it easy to switch LLMs

    Mixed · May 2026 · in “Ask HN: Which LLM are you using to evaluate your ideas?”

12. crazy complex prompt composers (langchain) that don't need to exist any more because the underlying model got so much better

    Criticizes · Apr 2026 · in “Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving”

13. langchain is stuck in innovator dilemma - it was built for gpt 3.5, or 4, it needs a different design for todays models

    Criticizes · Feb 2026 · in “Cord: Coordinating Trees of AI Agents”

14. agents are easy until you build a whole application using an API endpoint that OpenAI or LangChain decides to yank

    Criticizes · Nov 2025 · in “You should write an agent”

15. there used to be a whole complex apparatus around getting models to do chain of thought reasoning, e.g., LangChain

    Criticizes · Nov 2025 · in “Agent design is still hard”

16. it was evolving so fast that even the latest ai models didn't have even remotely up to date information on it

    Criticizes · Dec 2025 · in “Ruby 4.0.0 Released”

17. most of the functionality actually adds more code to use than simply writing my own direct API LLM calls

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

18. I am afraid it will all go in the LangChain direction with VC funding designs that are not yet ready

    Criticizes · Apr 2026 · in “Show HN: Atomic – Local-first, AI-augmented personal knowledge base”

19. Because it tries to solve non existent problems (prompt templates for example) and adds complexity

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

20. We have excellent "Actuators" (LangChain, Vector DBs) that execute actions, but they are blind to meaning.

    Mixed · Dec 2025

21. A lot of AI-adjacent libraries we've been using finally hit v1.0.0 this year, or creeping close to it

    Recommends · Dec 2025 · in “A new AI winter is coming?”

22. LangChain/Python is the least set up to take advantage of strong schemas even when you do have it.

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

23. langchain (despite its shortcomings) with its plethora of integrations makes things easier to do

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

24. Even Langchain, for all the (deserved) criticism it gets made this very clear very early

    Mixed · Jun 2026 · in “Apache Burr: Build reliable AI agents and applications”

25. Langchain is close but still feels a bit abstract and DIY.

    Mixed · Apr 2026 · in “Claude Managed Agents”

26. I think mine is coded to a higher standard than litelm.

    Recommends · Sep 2026 · in “Litelm: LiteLLM Without the Bloat”

27. I haven't touched LangChain in a long, long time

    Mixed · May 2026 · in “DeepSeek reasonix, engineered for low token cost during long running sessions”

28. now seems React agents are in LangChain

    Mixed · Nov 2025 · in “Agent design is still hard”

29. will retain usefulness into the future

    Recommends · Jul 2026 · in “Claude Cookbook”

30. the python version which is superior

    Recommends · Nov 2025 · in “Ask HN: What agentic framework do you use?”

31. it works pretty well on a 16G GPU

    Recommends · Jan 2026 · in “Ask HN: How are you doing RAG locally?”

32. they are inherently rigid

    Mixed · Feb 2026 · in “Cord: Coordinating Trees of AI Agents”

33. A critical LangChain core bug can turn a single LLM prompt into a secret-exfiltration or object-instantiation exploit.

    Criticizes · Jan 2026

34. it took our dev maybe six weeks of struggling with LangChain to approximate what Claude + MCP server can do today

    Criticizes · Sep 2026 · in “Muse Spark 1.3”

35. I have yet to see anything produced by multi-agent frameworks (langchain or bespoke works) that produced value.

    Criticizes · May 2026 · in “Multi-Agent is a snake oil”

36. ever after the LangChain situation I can see that what is adopted is always something that has a lot of flows

    Criticizes · Apr 2026 · in “Do I belong in tech anymore?”

37. if Anthropic themselves arent using it and using a prompt then what’s the big deal about langchain

    Criticizes · Mar 2026 · in “The Claude Code Source Leak: fake tools, frustration regexes, undercover mode”

38. It was just first to market, then after a short while most realized it's horrifically architected

    Criticizes · Mar 2026 · in “If Dspy is so great, why isn't anyone using it?”

39. There's ones that do not serve your use case, eg what was obviously day one regarding Langchain.

    Criticizes · Jun 2026 · in “Apache Burr: Build reliable AI agents and applications”

40. LangChain trying to be the Next.js/Vercel of agents but everyone recommending building your own

    Criticizes · Apr 2026 · in “Claude Managed Agents”