Jim Burtoft

[Jim Burtoft @JBurtoft]

@GroveCtyCollege I'm trying to call the registrar, the main line, or the admissions line and get a "call cannot be connected" message or a fast busy. Any problems you want to tell us about?

[Henry Yin✈️ICML @HenryYin_]

Most AI investing happens downstream of the frontier: a capability emerges, a category gets named, and capital rushes in.

But by the time a category earns a clean box on a market map, the best builders have usually been living in the messy version for months.

Agents. Reasoning. RL environments. World models. AI for Science. Recursive self-improvement.

I call this frontier proximity: the ability to see what is becoming possible before it becomes consensus.

My frontier proximity ladder:

L0 Wrapper: uses today’s models.
L1 Reactor: reacts fast to releases, but roadmap is downstream.
L2 Anticipator: builds for where capabilities are going.
L3 Native: depends on a non-obvious frontier bet.
L4 Shaper: helps move the frontier itself.

The point is not that every company needs to train models.

Apps can have high frontier proximity if they understand what models will make possible next.

Infra can have high frontier proximity if it knows what future agents, multimodal systems, robotics stacks, or scientific workflows will need.

That is why we’re launching MoE Capital.

MoE stands for Mixture of Experts.

The idea is simple: build an AI fund around people closest to the frontier: frontier researchers, technical founders, AI-native builders, and seasoned operators.

We don’t want to be another AI fund with a newsletter-level understanding of the frontier.

We want to build the AI fund closest to the frontier.

More in The Information: https://t.co/CXWJAy34zi

[Jim Burtoft @JBurtoft]

@HiltonHonors @HiltonHonors I am not saying it is against your terms and conditions. I am saying it is the wrong thing to do as a business. Don't fix it for me, fix it for everyone.

[Jim Burtoft @JBurtoft]

@HiltonHonors I understand late cancellation fees, but you don't give me point for the fees?!?

[Jim Burtoft @JBurtoft]

@_philschmid For me it has to be local because I’m using specialized hardware and libraries that they can’t (or won’t) replicate in a shared environment.

[Yoshitaka Haribara 針原 佳貴 @_hariby]

AWS Opens Applications for the 2025 Generative AI Accelerator to Support Startups Building Foundational AI
#GAIA
https://t.co/TejZ3ljH5o

[Jim Burtoft @JBurtoft]

I'll be one of the speakers at the AWS NYC Summit on Wednesday. CMP304 at 2:45PM. "Fine-tune Hugging Face LLMs using Amazon SageMaker and AWS Trainium". (We'll talk about Inferentia too!)

[Philipp Schmid @_philschmid]

Did you know you can deploy open LLMs on AWS Infernetia2? 👀 AWS Inferentia2 is a custom accelerator that is an alternative to sparse GPUs. 🔥

👉 https://t.co/E1gHyXFQlT

Learn how to deploy the Mixtral 8x7B model on @awscloud Inferentia2 using @huggingface Text Generation Inference and Amazon SageMaker.

TL;DR;
🚀 Setup: Step-by-step instructions to deploy Mixtral 8x7B using the Hugging Face LLM Inf2 Container on Amazon SageMaker.
🧑‍💻 Inference: Easily Chat with the model using the Messages API.
📊 Benchmarking: Throughput of 288.00 tokens/sec, inter-token latency of 10.67ms/token at 5 concurrent requests.
💵 Cost: Inferentia2 instance types range from $0.76 to $12.98 per hour on-demand.

Which model should we tackle next? 🤗