The real bottleneck in scaling AI isn't compute with Bassam Tabbara at Upbound episode artwork

EPISODE · Aug 25, 2026 · 50 MIN

The real bottleneck in scaling AI isn't compute with Bassam Tabbara at Upbound

from Hello, Agent!: The podcast at the intersection of data & agents · host Redpanda

In this episode, we talk with Bassam Tabbara, Founder and CEO at Upbound, Founder of Crossplane, and Founder of Modelplane, about what it takes to run AI inference at scale.Bassam walks us through his path from writing BASIC on a ZX Spectrum in Lebanon to building the early automation behind Hotmail at Microsoft, founding Crossplane, and then Modelplane: a new open-source fleet orchestrator built for running any model, on any engine, on any hardware.KEY TAKEAWAYS00:00 Alex introduces Bassam Tabbara, inventor of Rook and Creator of Crossplane, here to discuss his newest release, Modelplane.01:20 Bassam's first exposure to programming: teaching himself BASIC on a ZX Spectrum 48K as a kid in Lebanon.05:35 Control planes exist because human-in-the-loop failure response doesn't scale.07:57 Owning your own intelligence is fundamentally about data control, not just cost.11:05 Modelplane was built because Upbound's enterprise customers are already running inference across multiple Kubernetes clusters.13:45 The AI stack — models, serving engines, infrastructure, and accelerators — is tipping horizontal, with more vendor choice at every layer.22:35 The fleet-wide inference problem was unsolved in open source, so Upbound built Modelplane to close that gap.28:45 Modelplane separates concerns between platform teams and ML teams through two distinct APIs.39:35 AI is a workload, not a new platform, and that reframe changes how enterprises architect around it.48:25 Modelplane is an early, 0.1 release, and Bassam invites the open-source community to contribute.Thanks for listening to "Hello Agent!: The podcast at the intersection of data & agents." If you loved this episode, let us know with a 5-star review! Remember to subscribe so you don't miss an episode. To learn more about Redpanda, visit redpanda.comRESOURCES MENTIONEDModelPlane, Bassam's newly released fleet orchestrator for AI inference:modelplane.aiCrossplane, the open-source control plane framework Bassam created:https://www.crossplane.ioRook, Bassam's earlier open-source storage project:https://rook.io/Upbound website:https://www.upbound.io/Cloud Native Computing Foundation (CNCF), which now hosts Crossplane as a graduated project:https://www.cncf.io/vLLM, one of the open-source serving engines discussed:https://vllm.ai/SGLang, the other serving engine discussed:https://www.sglang.io/#RealTimeData #DataStreaming #Redpanda

Episode metadata supplied by the publisher feed · Published Aug 25, 2026

Embed this episode

Ready to play

The real bottleneck in scaling AI isn't compute with Bassam Tabbara at Upbound

0:00 50:44

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of Hello, Agent!: The podcast at the intersection of data & agents?

This episode is 50 minutes long.

When was this Hello, Agent!: The podcast at the intersection of data & agents episode published?

This episode was published on August 25, 2026.

Can I download this Hello, Agent!: The podcast at the intersection of data & agents episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!