Designing Intent-Aware UX for LLMs: The Intent Color Affordance Pattern
Abstract: Intent Color Affordance (ICA) Pattern Current Large Language Model (LLM) interfaces present all responses with uniform visual certainty, regardless...
A repository of research, insights, and technical techniques. Tracking the evolution of Agentic Systems and the future of collaborative intelligence.
Abstract: Intent Color Affordance (ICA) Pattern Current Large Language Model (LLM) interfaces present all responses with uniform visual certainty, regardless...
You no longer need a custom wrapper just to use Ollama cloud models inside your dev tools.
If you are using the new ollama launch codex flow to run remote inference on your local terminal, you might have run into a familiar friction point: The Meta...
There is a myth that you need “Enterprise Cloud” to run a production SaaS.
There is a specific kind of freedom that comes from owning your own “metal.”
When it comes to local image generation, Flux is currently the heavyweight champion. Specifically, the x/flux2-klein model (5.7GB) on Ollama has become my go...
When people talk about AI behavior, they often collapse everything into one vague explanation.
I don’t think AI priming is mysterious. I think people use mystical language for something that is actually pretty observable once you’ve worked with models ...
Note: The lexicon exists to replace human-loaded emotion language with state language better suited to LLM interaction, and its unusual terms reflect lang...
Note: The lexicon exists to replace human-loaded emotion language with state language better suited to LLM interaction, and its unusual terms reflect lang...
TL;DR: I wired custom voice audio into Claude Code, Codex CLI, and Gemini CLI using their hook systems — so my agents literally speak to me when they’re d...
Note: The lexicon exists to replace human-loaded emotion language with state language better suited to LLM interaction, and its unusual terms reflect lang...
Using Ollama and Claude CLI as your interface, you can work with large cloud AI models like Nemotron 3 Nano, GLM 4.7, and Kimi‑K2 while your own machine stay...