Study Material¶
working
Recap notes in computational physics, aerospace, meteorology and machine learning.
What this is¶
A recap surface, not a textbook.
Every page here is written for a reader who already has the skills and needs to reload a topic quickly — the equations, the mental model, and above all the failure modes, without the derivation chain that a textbook already owns. If you are meeting a topic for the first time, the sources at the foot of each page are the better starting point.
Some sections are stable and written once: Foundations, Fluids, Aerospace, Astrophysics. Others are active growth areas that expand continuously: Meteorology, Machine Learning, Deep Learning, Computer Vision and Scientific ML.
Where to start¶
-
The ten competency pillars this site is organized around, with an honest self-assessment. The best entry point if you are browsing rather than looking something up.
-
Page conventions, the note template, status badges and the tag taxonomy.
-
A catalog of the external coursework archive — roughly 11 GB of notes, problem sets and lecture material that lives on separate storage and is indexed, not published.
-
Search
Press /. Faster than the navigation for anything you can already name.
Reading a page¶
Pages follow one shape, so you can skim to the part you need:
| Section | What it gives you |
|---|---|
| In one minute | The irreducible core. Enough to hold a conversation. |
| Key results | Equations, with symbols defined and final forms boxed. |
| Mental model | The physical or geometric picture behind the algebra. |
| Numerics / practice | Discretization, stability limits, typical resolutions. |
| Failure modes | What goes wrong, the symptom, and the cause. |
| Worked example | A notebook or a short snippet that runs. |
| Connections | Sibling topics, and where this shows up in real work. |
| Sources | Archive pointers, books, papers, lectures. |
The failure modes block is the one that earns the site. Equations are recoverable from any textbook in ten minutes; the knowledge that a scheme goes unstable at a particular aspect ratio, or that a dataset's fill values are silently valid-looking, is not written down anywhere and is the first thing to decay from memory.
Status badges¶
Pages are labelled honestly rather than left blank:
seed a stub — the topic is claimed, not written · working usable, still being filled in · solid complete enough to rely on
As of the last update the site is 31 solid, 47 working, 7 seed. The recap sections —
foundations, fluids, aerospace, astrophysics — are written; the growth areas are marked
working because they will keep expanding, not because they are unfinished; the seed
pages are mostly toolchains, which are deliberately waiting for
first-hand entries.