X postMahaxOpinion
Building multi-agent or multi-step LLM systems as linear pipelines creates bottlenecks and brittle failure points. Applying graph engineering principles allows developers to run independent AI operations in parallel and structure workflows based on strict data requirements.
Mahax argues that while creating individual agents has become accessible, orchestrating the execution graph behind them is the real differentiator. Public replies point out that the referenced lecture is Andrej Karpathy's Stanford CS25 talk from 2023 on transformers and attention as graph message-passing. Mahax translates these conceptual fundamentals into concrete workflow design rules, such as pruning fake dependencies and using fan-out patterns.
Mahax
@Mahaximus_ · 10 September 2026
Andrej Karpathy: "Prompting is going away. Delete everything, keep Graph." In 1 hour he shows how to build Graphs, and why it's the only thing that will be left standing at the end. Anyone can build an agent, almost no one builds the Graph that runs them. Watch it, then read t.co/tZpu4Byhy4 t.co/I3b2Rt1Wvv
919 likes18 replies2.4k bookmarks240.8k views
View on XSummary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.
Similar in content, found by the AI search.
X post:nicco
A post by nicco highlights remarks by Andrej Karpathy asserting that prompting is an ephemeral interface and that graph architectures represent the true architectural layer for agent workflows. It pairs this with a practical guide by seeco on replacing sequential agent chains with explicit graphs.
529bookmarks58.5kviews
X post:Mahax
Citing Andrew Ng, Mahax highlights the shift from single-shot prompting to agentic loops and graphs. An accompanying guide outlines the architecture of AI agents, self-evaluating feedback loops, and graph-based execution pipelines using Claude.
11.8kbookmarks1mviews
X post:rvaniaaa
rvaniaaa contends that the primary bottleneck in multi-agent workflows is geometry rather than model intelligence. Switching from sequential agent chains to graph-based architectures allows parallel execution and substantially reduces overall execution time.
2.9kbookmarks233.5kviews
X post:nicco
Citing a Stanford lecture by Andrew Ng, author nicco highlights why zero-shot prompt engineering is reaching its limits and being replaced by autonomous execution graphs, supported by a guide on moving away from linear agent chains.
275bookmarks25.1kviews
X post:Codez
An X post by Codez quotes Claude Code creator Boris Cherny claiming that agent systems will shift from manual prompting to graphs and loops. The post introduces a guide outlining graph-based architectures for multi-agent execution.
5.4kbookmarks549.2kviews
X post:Mahax
An X post by Mahax highlights a one-hour lecture by Jeff Dean spanning LLM basics to coordinating 100 AI agents. The post also references an accompanying guide on graph engineering using Claude.
3.8kbookmarks273.9kviews
Parts of this content were created with the assistance of AI.
