A quiet archive of essays, architectural post-mortems, and reflections on building software.
Most tools optimize for features and organizational overhead, adding friction rather than removing it.
What constraints taught me about model architecture, lightweight NLU, and optimization during the SAM era.
Why engineering elegance alone isn't enough, and why understanding user context is what actually builds value.
The recurring pattern across my work and the thesis that software should adapt to people, not the other way around.