Buying Tokens vs Building an AI Factory with Sadeepa Wijesekara
EPISODE 785
|
NOVEMBER
17,
2026
Watch
Follow
Share
About this Episode
An AI agent can make many model calls to finish a single job. Some need frontier reasoning. Others can run on an open-weight model, and some involve data that has to stay in-house. How do you decide which inference to buy and which to run yourself?
Sadeepa Wijesekara, who leads engineering for HPE Private Cloud AI, joins Sam Charrington to discuss how enterprises make those decisions. They get into what it takes to see the cost of individual calls, how teams route work between hosted and local models, and when running your own GPUs makes economic sense. Sadeepa shares how HPE routes coding requests through a single internal endpoint that can send work to a frontier provider or serve it locally.
About the Guest
Thanks to our sponsor HPE
Thanks to HPE for supporting the show and sponsoring our AI Economics series.
As AI workloads scale, every token carries an infrastructure cost. HPE Private Cloud AI helps enterprises improve those economics by reducing wasted tokens and getting more value from their GPU investments. Organizations can run the right models on the right resources while maintaining control over cost, security, and governance.
To learn more about HPE Private Cloud AI, visit twimlai.com/go/hpe.
And for a look at how HPE is managing its own AI economics, check out my TWIML Briefing Room conversation with HPE Chief Data and AI Officer Dan Waibel. Link in the description.
Resources
- HPE Private Cloud AI
- Apple is reportedly cutting budgets for its internal AI usage — TechCrunch
- Tokenomics Foundation
- NVIDIA OpenShell
- Kimi models — Moonshot AI
- GLM models — Z.ai
- NVIDIA H200 GPU
- NVIDIA RTX PRO 6000 Blackwell Server Edition
- NVIDIA Confidential Computing
- National Institute of Standards and Technology (NIST)
- Federal Information Processing Standards (FIPS) — Compliance FAQs
.png)