Álvaro Bartolomé

alvarobartt

Tech Lead & Engineer @ Hugging Face

Open models, inference, and cloud

Writing

/goal GLM 5.2 on Microsoft Foundry with AMD MI300X

Agents on AWS SageMaker AI with Strands Agents

Choosing wisely

Rust is strict, whether you like it or not

Estimate inference memory requirements w/ hf-mem

How to read and parse JSON with Zig 0.13

About

Tech Lead & Engineer at Hugging Face , working on inference + cloud. Currently, tech lead for cloud partners (Google Cloud, Microsoft Azure, AWS, Dell), and core maintainer of Text Embeddings Inference, making embedding models fast and accessible at scale.

Also interested in post-training, synthetic data generation, and LLM distillation, particularly in how these techniques can push model quality and efficiency further.

I also write Rust, Python, C, and Zig; with strong interest in low-level programming, distributed systems, infrastructure, and kernel-level optimization.