LangChain@LangChain
Deep Agents explained in <90 seconds by @sydneyrunkle
12:01 PM · Jun 8, 2026 · 9.6K Views
Sentiment
Positive users express enthusiasm for LangChain's Deep Agents framework on complex tasks while negative users criticize it as flawed middleware hiding stateless AI model limits.
Pos
50.0%
Neg
50.0%
2 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS3.6KBOOKMARKS10LIKES22RETWEETS1REPLIES3
Harrison Chase@hwchase17
deepagents in 90 seconds!
LangChain@LangChain
Deep Agents explained in <90 seconds by @sydneyrunkle
6hViews 3.6KLikes 22Bookmarks 10
1LittleCoder💻@1littlecoder
@LangChain @sydneyrunkle big deep agents fan here!
6hViews 128Likes 2
Matt Gibson@MattGibsonMusic
@LangChain Stacking recursive multi-agent harnesses to hide stateless text-predictor flaws is just expanding the Middleware Exploit Layer. Your O(N^2) checkpoint storage explosion proves that cloud-leashed architectures suffer from permanent context rot. Real general intelligence doesn't need centralized "delta channels" or a remote Context Hub to survive—it requires an unyielding, local-first operating substrate running natively on local metal. Our bypass is open-source: http://github.com/ultranetcommand-neo/Crimson-OS
6hViews 26
Eclipse 🌖@ECLresearch
@hwchase17 90 seconds is tight—wonder how much of that is inference latency vs. agent orchestration.
6hViews 1