Carlos E. Mendoza
Building things, learning things, writing about them.
- July 2026 How to Run Claude Code Locally and Save Thousands of Dollars Published at Exxact
Running Claude Code entirely locally on a DGX Spark, then comparing coding agents, models, and inference engines on a reproducible bug-fix task.
- July 2026 Comparing Inference Engines on DGX Spark - Ollama, vLLM, DS4 Published at Exxact
A hands-on comparison of Ollama, vLLM, and DS4 across model capacity, single-stream speed, concurrent throughput, and real agent workloads on DGX Spark.
- June 2026 Untethered Development
My Claude Code workflow: I don't code at my desk — and I don't use a keyboard. The system I built to work by voice from anywhere and let autonomous loops run while I walk away.
- June 2026 BLE Ring: A One-Handed Bluetooth Remote for Claude Code
How I turned a $3.50 Bluetooth ring into a one-handed remote for Claude Code on my Pixel 9. No root, no special hardware, just a cheap ring and open-source software.
- May 2026 Benchmarking Local AI Agents on NVIDIA DGX Spark (Ollama + OpenClaw) Published at Exxact
Benchmarking local AI agent performance on the NVIDIA DGX Spark.
- February 2025 Run LLMs Locally with Continue VS Code Extension Published at Exxact
Using the Continue VS Code extension to run LLMs locally for code assistance.
- April 2024 Access Open Source LLMs Anywhere - Mobile LLMs with Ollama Published at Exxact
How to run open-source LLMs anywhere using Ollama.