← Front page

Practical AI · Thursday, July 9, 2026

Agents as Unrolled Graphs: Rethinking Agentic Workflows

Hamza Tahir explains that all software, including AI agents, can be viewed as a graph of execution steps. While traditional ML pipelines were more deterministic, agents involve dynamic, potentially cyclical workflows. ZenML's evolution to a dynamic mode addresses this by allowing graphs to be defined in real-time.

companyZenML

The tape

3 quotes
Yeah, I mean, like, uh, all software is essentially a graph, right? When you write code, you have if this, else that, execute this, execute that. That's like a sequence of steps you're doing in a graph.
Hamza Tahir
So, uh, it's, it's where we have to reinvent ourselves as well. So, like back in like 2023, users started hacking our pipeline engine to run agents, like dynamic steps, conditional branching, state management, retryable tools, workarounds.
Hamza Tahir
Um, and, uh, we introduced a new dynamic mode. And I think the key difference here is really, as you said, everything is a workflow, um, everything is an an agent, in my opinion, is just an unrolled graph.
Hamza Tahir
Heard on Practical AI — “Building Durable AI Agents, published Thursday, July 9, 2026. Heardvine summarizes and quotes with attribution and timestamps, and links to the original everywhere.
Transcribed via Gemini audio transcription · $0.05
Agents as Unrolled Graphs: Rethinking Agentic Workflows — Heardvine