Skip to content

Undergraduate Guide

This page is a guide for reading the existing handbook — it is not a new course and it does not contain the main technical content. The handbook already covers the physics, the methods and the code. What this undergraduate layer adds, sitting on top of the handbook, is a set of reading strategies, vocabulary help, formula help, gentle learning paths, small projects, and self-checks. Think of it as a study companion that walks beside the existing book rather than a second book that competes with it.

If you read nothing else, read this: you are allowed to go slowly, and you are not expected to understand everything on the first pass.

Who this guide is for

This guide is for a normal undergraduate in materials science, physics, chemistry, or mechanical or chemical engineering. You have seen calculus, vectors and matrices, and some first-year physics, but you may not feel confident with them yet. You may have written a little Python and used a command line once or twice, but you would not call yourself a programmer. You probably have little or no research experience. None of that is a problem. The handbook was written to be learnable, and this layer exists to make the climb less steep.

What you are expected to know

  • Maths: A-level or first-year-university level. You should recognise differentiation and integration, vectors, and basic matrix operations, even if you need to look up the details.
  • Physics: A-level or first-year level — forces, energy, waves, and a first taste of atoms.
  • Python: enough to read a short script, run it, and change a number. You do not need to write programs from scratch.
  • Command line: enough to open a terminal, change directory, and run a command someone gives you.

What you are NOT expected to know

  • Any research experience, or how to read a scientific paper.
  • Advanced quantum mechanics. A first-year picture of wavefunctions and energy levels is plenty to begin.
  • The internal machinery of density functional theory (DFT). You will meet DFT gently in Chapter 5 (DFT); you do not need to know how it works on day one.
  • Machine learning. The later chapters on machine-learned potentials and graph neural networks build up from the start.

It is normal to feel tired

Technical material is dense. A single paragraph can introduce a new symbol, a new acronym and a new idea all at once, and that is genuinely tiring to absorb. Feeling tired is not a sign that you are not clever enough — it is the expected cost of real learning, and every researcher has felt it. When a section starts to feel like too much, slow down deliberately:

  • Re-read the plain-language part at the top of the section before the formal version. The intuition is meant to carry you.
  • Write the one-sentence summary of the section in your own words. If you cannot, you have found exactly what to re-read.
  • Look terms up in the beginner glossary, which explains words slowly. (There is also a terser Appendix C glossary in the main book for quick reference once you know a term.)
  • Do one worked example by hand before moving on — even a tiny one. Following the numbers yourself fixes an idea far better than reading does.

Rest, then return. Difficult sections almost always feel easier the second time.

The three learning layers

The handbook can be read at three depths. You do not have to read all three on your first pass.

  • Layer 1 — Undergraduate Core. The essential ideas, the intuition, toy examples, basic by-hand calculations, and short pieces of code you can run. This is where you should spend your first read. If you understand Layer 1, you understand what the chapter is about.
  • Layer 2 — Advanced Notes. More formal derivations, deeper theory, and technical caveats. This is genuinely useful, but it is safe to skim on a first pass and come back to later when the core idea has settled.
  • Layer 3 — Research Path. Real workflows, modern methods, capstone-style projects, and the judgement needed to read the literature. This is where the field actually lives, but it is optional on a first read and assumes Layer 1 is comfortable.

Aim for Layer 1 first

On a first read of any chapter, your only goal is Layer 1. Treat Layers 2 and 3 as a return visit. Trying to absorb all three at once is the fastest way to feel overwhelmed.

Before you move on

Before leaving a section, check that you can honestly tick each box. If you cannot tick one, that is your signal for what to revisit — not a reason to give up.

  • I can state the main idea of this section in one sentence.
  • I can say what physical system is being discussed (a single atom? a crystal? a box of moving atoms?).
  • I can define each new piece of vocabulary that appeared.
  • I can point to the key equation and say what it calculates.
  • I can explain the simplest example given.
  • I can run the code, or at least follow what each line does.
  • I can answer the "Check Yourself" questions in the chapter.

Where to go next

Once you are comfortable with how to read the handbook, move on to the rest of this undergraduate layer:

It is also worth reading two pages from the main Start-Here section: