AI-Assisted Llama.cpp Optimization
How Claude Sonnet 3.7 Maximized llama.cpp Performance For My Hardware
Unloading Ollama Models Instantly
A small alias to help clear your GPU's memory
Through the Looking-Glass
Peeking into private and airgapped networks through the use of a proxy server
Fully automated and self-hosted production builds with GitLab
Zero cloud reliance while building and deploy my applications.
Work
- Company
- Brightspot
- Roles
- Director, Engineering (2025 - Present)Solutions Architect (2023 - 2025)Staff Software Engineer (2022 - 2023)Primary Oncall Agent (2021 - 2023)Senior Software Engineer (2020 - 2022)Software Engineer (2016 - 2020)