---
title: "Rohit on X: Frameworks and Composability Insights"
url: https://stacklist.com/card/6546962f-8785-4908-919e-56d683af6d9f
source_url: "https://x.com/rohit4verse/status/2085492848103915977?s=12"
stack: https://stacklist.com/c/technology/stack/1659549d-373d-4391-ba12-5a14d40c19ed
summary: "Rohit discusses a composable three-layer architecture for AI agents (harness, framework, runtime) as a unified approach to framework selection. The post references a comparison of Deep Agents, LangChain, and LangGraph, highlighting their distinct approaches to building agents."
tags: "agent-frameworks, langchain, langgraph, composable-architecture, ai-development, framework-comparison"
key_entities: "Rohit (person), Sydney Runkle (person), LangChain (technology), LangGraph (technology), Deep Agents (technology), Claude (technology), composable-architecture (concept)"
classification: "analysis"
content_hash: "sha256:914a8751c4fffab9b014ead01c6d7f48adf67689585822e699b8af2b2c3fe2d0"
acp_version: "0.2"
token_counts_approximate: 137
visibility: public
agent_accessible: true
status: "final"
---

# Rohit on X: Frameworks and Composability Insights

Rohit @rohit4verse RIP to every "which framework should I use" thread. harness on top, framework in the middle, runtime at the floor. one stack, three heights, fully composable. paste these four images into claude and ask which layer your agent belongs on. Sydney Runkle @sydneyrunkle 11h Article Deep Agents vs LangChain vs LangGraph Deep Agents, LangChain, and LangGraph each offer distinct approaches to building agents. In this post, we cover the key distinctions between our open source frameworks and when you should reach for...