# Alex Finn: Running 24/7 Local AI on His Own Hardware A solopreneur optimizing hardware with local AI models. - **Source:** [Lenny's Newsletter](https://www.lennysnewsletter.com) - **Published:** 2026-07-13 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/b0a9109f-ae03-47d1-8b8a-1452a2a256a8 ## Who is Alex Finn? Alex Finn is the founder of Vibe Code Academy, a platform teaching AI tools, and a YouTuber. He runs local AI models on his own hardware, including Mac Studio, DGX Spark, and a custom RTX 5090 setup. ## Hardware Setup Alex uses three Mac Studio 512 GB machines, a DGX Spark, and a custom RTX 5090. These devices are managed through a dashboard he developed himself. He spent five months optimizing which AI models to run on each machine. ## Tailscale and Network Management Alex emphasizes the benefits of using Tailscale, even on a single machine. This tool allows him to manage his entire hardware fleet through one agent. It also integrates with Claude Code loops for seamless operation. ## Models and Task Allocation Alex optimizes models like GLM 5.2, Qwen 3.6, and Ornith 1.0 for specific tasks. He assigns tasks based on each model's strengths, showcasing how local inference reduces cloud costs significantly. ## OpenClaw vs. Hermes Comparison Alex compares OpenClaw and Hermes models. OpenClaw is better suited for specific tasks, while Hermes offers broader applications. He runs five agents in total, detailing the pros and cons of each. ## The Future of Local AI Alex's work demonstrates how local AI can be more economical and secure compared to cloud-based solutions. This approach holds great potential, especially for solopreneurs and small teams. ## Sources covering this story - https://www.lennysnewsletter.com/p/this-solo-builder-runs-247-local — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._