by joacod
AudioBits: Turn Interactions Into Sound
Why I built AudioBits, a small procedural audio library for sounds that react to your application, with editable recipes and tools for coding agents.
2026
12 posts published in 2026.
by joacod
Nerdearla Reminded Me Why Showing Up Matters
My experience at Nerdearla, why the open-source conversations stayed with me, and how meeting curious people makes networking happen naturally.
by joacod
The Blog Was the Shelf, the Lab Is the Workbench
Why I created joacod Lab for interactive experiments, what its AI-first stack has taught me, and why building a small place for unfinished work is worth it.
by joacod
I Built SelfContext So I Could Stop Re-explaining Myself to AI
SelfContext is my local, portable take on Karpathy's LLM wiki, pointed at personal context instead of research notes, so useful AI conversations can continue from files I own.
by joacod
AI Gets More Useful When You Build the Workflow Around It
The biggest productivity gains from AI come from the context, skills, and workflows you build around it, because each useful project makes the next one easier.
by joacod
A $10K Private AI Is Starting to Look Reasonable
Local inference is moving beyond offline chat. A MacBook demo shows how a model becomes a private OpenAI-compatible service, and why that architecture could scale to serious on-premise AI.
by joacod
Small AI and the Return of Better Software
Small local AI models are pushing software back toward faster, cheaper, more private, and more focused product experiences.
by joacod
What I Learned Building My Own AI Harness
After building nano-harness, I have a clearer view of AI assistants, local-first workflows, architecture, and why building the small version yourself still teaches more than reading the hype.
by joacodPart 2 of 2
Context Is Not Memory, It Needs an Engine
Context is one of the real bottlenecks in agent-native engineering, and treating it like passive memory is not enough. It needs a system that can manage and surface it properly.
by joacodPart 1 of 2
Human Code Review Is Not the Last Frontier
Human code review is not the final barrier to autonomous software engineering. The deeper challenge is judgment, context, and the messy reality of real codebases.
by joacod
My Personal Blog Is Finally Live
After two years publishing on Dev.to and Medium, I launched my own blog and rebuilt joacod.com with Astro and AI-assisted workflows.
by joacod
AI in Corporate, Navigating a Sea of Hot Air
I see several problems in corporate environments and in how AI is being used to speed up...