How developer hardware works: standards, specs and setups

DevWorkstation Notes explains how developer hardware works, from processors, memory and storage to monitors, desk equipment and home networks. Every article is built from the public documentation of standards bodies, government programs and manufacturers, and cites its sources in the text.

Laptops and desktops

Machines for programming, AI and data science work.

Local AI and GPUs

Memory, compute and the cost of running models yourself.

  • How much GPU memory local LLMs needLocal LLM GPU memory needs depend on parameter count and precision; 2 GB per billion parameter rule at bfloat16 helps estimate, with quantization and context.
  • CUDA compute capability explainedA guide to CUDA compute capability, how to find a GPU's version, and why it matters for CUDA, PyTorch, and machine learning.
  • When renting a GPU beats buying oneHow to decide between renting and buying GPUs for local AI workloads, including compute capability, benchmarks, memory, and software.

PC parts

Processors, memory, storage, power and cooling.

Desk setup

Monitors, input devices, chairs, cameras and audio.

Homelab and networking

Storage servers, backup power and the network around them.