# Optimizing Hugging Face Carbon Models on AWS Trainium2 with NxD Inference

Page: https://stenobird.com/podcast/the-business-compass-llc-podcasts-7078188/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference
Text version: https://stenobird.com/podcast/the-business-compass-llc-podcasts-7078188/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference.md
Podcast: [The Business Compass LLC Podcasts](https://stenobird.com/podcast/the-business-compass-llc-podcasts-7078188)
Published: 2026-06-11T05:19:33+00:00
Episode link: https://podcast.businesscompassllc.com/e/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference/
Audio file: https://mcdn.podbean.com/mf/web/pgu3m64yzxei7nkd/0ff972d6-c125-461a-87ae-cad77975f225.mp3
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/the-business-compass-llc-podcasts-7078188/episodes/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference
Duration seconds: 849

## Resource

Running large language models in production means dealing with sky-high costs and frustrating latency issues. AWS Trainium2 changes the game by offering specialized hardware designed for machine learning model scaling, while Hugging Face Carbon models provide the efficiency modern applications demand.

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/the-business-compass-llc-podcasts-7078188/episodes/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/the-business-compass-llc-podcasts-7078188/optimizing-hugging-face-carbon-models-on-aws-trainium2-with-nxd-inference.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.