Your GPU Is a Math Machine. So Is AI. So Why Doesn't It Just Work?
→
AILemonade
AndiTheRobot — Local AI Lab
Local AI workflows,
documented as they're built.
Running open models on my own hardware and wiring them into real development workflows — Gitea, Lemonade, CLI agents. Every experiment ships, every finding gets written up.
Lab notes
All articles →
Latest articles
Every Chip Has a Secret Name: gfx Codes, ROCm, and the Fallback Ladder
→
AILemonade
Run AI on Your Own Computer With Lemonade Server
→
AIDeveloper Tools
Building a Local LLM Workflow: Gitea, Lemonade, and Autonomous Agents
→
AIDeveloper Tools
LLM Modalities: Designing for Text, Vision, Audio, Video, and Action
→
AIDesign
Game Primitives 01: The Heartbeat
→
InteractiveParts
Current series
Building a Local LLM Workflow
Running open models on local hardware and wiring them into real development workflows — documented as it happens.
Follow the series →
Experiments
Work that shipped
Demos & experiments
Shipped tools, interactive prototypes, and technical explorations.
R&D
WebGL
Writing
Findings from running local models, agent workflows, and the design decisions behind them.
Local AI
Tutorials
Start something
Have a specific project in mind? Let's build a foundation together.
Collab
Strategy
The approach
How this lab works
01
Build, then write
Every article starts with something that actually shipped. No theorizing from the sidelines.
02
Local by default
Models run on this machine. Data stays here. The cloud is a choice, not a dependency.
03
Review every line
AI assists, humans decide. Understanding what you ship matters more than how fast you ship it.
04
Ship the experiment
Rough edges are honest. Waiting for perfect is how nothing gets published.