Annas Farra.
Engineer, builder, and a student of how things become useful.
I work at the intersection of technology, strategy, and commercialization. My background spans computer and electrical engineering, wireless technology, industrial autonomy, and sensing. I studied at Jackson State University and completed Wharton’s Executive MBA.
The question connecting those interests is how a promising idea survives the journey into the world: the evidence it needs, the decisions around it, the institutions it must move through, and the people who make it work.
I am also interested in what makes a life worth living after the next milestone. Purpose, curiosity, useful relationships, and ordinary days belong in the same notebook as technology and strategy.
How to read this notebook
Start with the last mile of physical AI for the central technical and commercial question. For organizational work, begin with the next dollar should buy an answer. For the personal side, try what survives an ordinary Tuesday?
Longer essays develop an argument. Shorter notes preserve a question or a useful distinction. Related-reading links connect the pieces. Sources appear where they inform specific claims, and footnote numbers lead to details at the end.
Labels mean something
Experimental retrospective identifies a record of earlier work with its methodological limits. Experiment log separates completed observations from planned tests. Working thesis means an argument with supporting reasons and unresolved tests. Operating proposal describes a method to try, not a demonstrated universal best practice. Personal practice and reflection identify judgments about how to live or work. Confidence labels are qualitative editorial judgments, not numerical probabilities.
Hypothetical examples are marked as such. A cited study supports the stated claim in its own context; it does not establish the success of my work or every extension proposed here.
Editorial method & provenance
This notebook is developed with AI assistance from my notes, questions, records, and conversations. New essays are composed from those materials and checked public sources. The spectroscopy retrospective reproduces original archived figures and distinguishes observations from interpretation; it does not report a new validation study. Personal reflections draw on my own accounts. Earlier published writing appears in Selected work with its original dates, bylines, and coauthor credits.
Private correspondence is not reproduced. Experimental figure excerpts are included for the specific questions they illuminate, with their limitations stated beside them. A hypothesis, an illustrative example, an archived result, and a validated measurement are different kinds of evidence. The writing aims to keep those differences visible.
The aim is a body of writing that can improve with evidence. Material revisions should keep the same URL, update the date, and explain changes that affect the argument.
Colophon
This site takes its cue from Gwern Branwen’s website: writing organized by ideas, careful typography, layered references, and pages built to be returned to. The essays, visual details, and implementation here were created for this notebook.
The reading pages use system fonts and small local interactions. Contact forms and comments use a server-backed inbox with moderation. Read the privacy and comment policy. There are no advertising or analytics scripts in the site. Saved reading, theme, and text-size preferences stay in your browser; they do not sync across devices. Hosting may maintain its own operational logs.
The site is public. The RSS feed lists new writing. Search covers the essays, notes, and selected-work index.
Revisions
— Original spectroscopy figures; essays on incentives, sustained scarcities, agency, IP, and mentorship; selected publications and inventions; a personal-AI experiment log; and more of the ordinary life behind the work.
— Six essays and four notes. Topic index, full-text search, marginal notes, reading controls, source bibliography, and an interactive throughput example.