Humanity’s Cosmic Endowment

The cosmic endowment is the matter and energy a future civilization could reach. Who should inherit it, and what might we owe conscious minds we create?
Notes on software, AI tools, books, and life outside work.

The cosmic endowment is the matter and energy a future civilization could reach. Who should inherit it, and what might we owe conscious minds we create?

A Labor Day weekend backpacking trip from Granite Lake to Upper Velma Lake, with a new sleep setup and more food than we needed.

We flew to Las Vegas, drove into southern Utah, and hiked around Zion, Escalante, and Bryce Canyon.

I’m using expensive agent sessions to refine the instructions that tell Codex how I want the work done.

My first week in San Francisco: paperwork, an unfurnished apartment, two cats, Bay Trail runs, and waiting for furniture.

A browser-based load-flow solver and Mandelbrot explorer let me revisit power-system analysis, numerical precision, and the limits of small client-side tools.

A concrete way to understand dropout in deep learning: sampled subnetworks with shared weights, standing in for a much more expensive ensemble.

Notes on the 11-inch M3 iPad Air for reading, handwriting, and remote access without turning it into a laptop replacement.

This site generates responsive image variants and their lookup metadata together, so components and Markdown do not have to reconstruct filenames.

A rough internal-tool prototype could let people test their understanding of a workflow before committing engineering time to it.

Making every part myself matters less to me when AI tools still let me express my ideas and taste.

Notes from Chapter 6 of Deep Learning on representational capacity, low-rank parameterizations, and the choice of activation function.

Part I of Deep Learning reviews the linear algebra, probability, and numerical computation that later chapters build on.

Agents can implement tasks in parallel; reviewing plans, checking edge cases, and coordinating the work still fall to me.

How repo-backed Markdown becomes static HTML without a CMS, database, or runtime admin surface.
Follow.it sends the emails and includes an unsubscribe link.