GPUs para IA
Guía técnica sobre las arquitecturas de GPU de NVIDIA, los formatos de precisión que cada una acelera en hardware, y cómo eso decide qué modelos puedes cargar, a qué velocidad y con cuánta VRAM.
Todas las entradas
15 artículos
GPUs for AI
A technical guide to NVIDIA's GPU architectures, the precision formats each one accelerates in hardware, and how that decides which models you can load, at what speed, and with how much VRAM.
Cómo generar imágenes profesionales en Helmcode: la guía completa de prompting
La diferencia entre una imagen con 'cara de IA' y una profesional no es el modelo: es el prompt. Guía completa para generar imágenes increíbles en Helmcode.
How to Generate Professional Images in Helmcode: The Complete Prompting Guide
The difference between an image that screams 'AI' and a professional one isn't the model, it's the prompt. The complete guide to generating great images in Helmcode.
hello, world: borja pérez here_
Nueva incorporación en Helmcode: comunicación, marketing y comunidad. Quién soy, qué vengo a hacer y qué planes tenemos con la comunidad de NaN.
hello, world: borja pérez here_
New hire at Helmcode: communication, marketing and community. Who I am, what I'm here to do, and what we're planning for the NaN community.
Qwen3.6 en RTX PRO 6000 a máxima escala
Cómo pasamos Qwen3.6-35B a NVFP4 en una sola RTX PRO 6000, por qué se caía, y el fix de una línea que resultó ser un bug de cuDNN, no de vLLM.
Qwen3.6 on an RTX PRO 6000 at full scale
How we moved Qwen3.6-35B to NVFP4 on a single RTX PRO 6000, why it kept crashing, and the one-line fix that turned out to be a cuDNN bug, not a vLLM bug.
NaN: the community's first month in numbers
117 billion tokens, 3.68 million requests, 21 countries, and 99.98% uptime. NaN is a community of builders with its own inference infrastructure and a private platform to deploy apps and agents.
AI for Dummies - Part 1
In this post we'll learn what parameters and quantization are, so we can figure out how much space AI models take up.
How to set up an inference server to serve AI models
In this post I'll walk you through how the community's inference servers are set up: the hardware we use, the stack we run, and the models we serve.
Claude Code for DevOps
I've spent several hours over several days documenting and optimizing my entire local environment so I can "mechanize" the work I do every day managing infrastructure for multiple startups.
Clawd/Molbot/OpenClaw: Reflections and lessons learned
This post isn't meant to be a guide on how to use Clawd, but rather a look at how we're rolling it out at Helmcode to have an AI Agent that helps us with our day-to-day work managing the Cloud infrastructure of multiple startups.
Kubernetes for devs (1/2)
Kubernetes is one of the most widely used infrastructure tools among companies, and it has become the standard for running containerized applications at scale all over the world.
Moving Down from Serverless to Kubernetes
Before we start, a bit of context. The infrastructure is hosted on AWS and the architecture was based on Serverless services: