Study tips for computer science students: beyond "just code more"

Computer science · 7 min read

By Raul Blanco · Updated 12 July 2026

Every CS student eventually receives the same advice: just build things. It is completely true and completely useless, like telling someone the way to get rich is to have money. Here is the version with actual instructions.

Computer science is really two degrees sharing a student card. One is a craft you learn with your hands, through projects and assignments and bugs. The other is a maths degree that occasionally mentions computers. They need different study methods, and most struggling CS students are using the wrong one for the course in front of them.

Escape the tutorial trap

Watching someone else code is television with extra steps. Your brain nods along, everything makes sense, and three days later you cannot write a for loop that filters a list. This is not a personal failing, it is how passive input works: recognition builds fast, ability does not.

The fix is one rule: never finish a tutorial without rebuilding something from it with the tutorial closed. Watch the section, close the tab, write it from memory, then compare. The version you write will be worse, and the ten minutes you spend finding out why will teach you more than the hour of watching did.

Coding assignments: budget for the debugging tax

Every assignment has two halves: writing the code, and staring at the code wondering why it hates you. The second half is the debugging tax, it is usually about half your total time, and no amount of experience makes it zero. Estimate how long the assignment will take, then double it. That doubled number is the real one.

This is also the real argument for starting early. Not to finish early, nobody finishes early. You start the day it is assigned so you can find out where it is hard while help still exists. The worst place to discover that the assignment is secretly a graph problem is 11 p.m. the night before, when the lab is closed and the forum has gone quiet. If getting started is the wall, the ten-minute contract in our procrastination guide was built for exactly this.

Data structures and algorithms: paper first

DSA courses punish passive reading harder than any other subject. Understanding a hash map when someone explains it is trivial. Implementing one at a whiteboard, or picking the right structure in an exam, is a different muscle. Train it directly:

• Implement from memory. Once per structure: close the notes, write it, run it, fix it. This is active recall in its natural habitat.
• Explain the complexity out loud. Not "it's O(log n)" but why it is O(log n), to a wall, a friend, or a rubber duck.
• Do blank-paper traces. Take a small input and walk it through the algorithm by hand. Exams love this, and it is where hand-waving goes to die.

Theory courses are maths courses in a hoodie

Discrete maths, automata, complexity theory: these do not respond to coding energy. Reading a proof and nodding is the same trap as watching a tutorial. The only way through is doing problems by hand, getting them wrong, and reading the solution only after a real attempt. Twenty problems half-understood through struggle beat a hundred pages of understood-while-reading.

Time-box your debugging

There is a special trance where you try the same three fixes in rotation for two hours, understanding less each cycle. Install a rule before you need it: 45 minutes stuck means stop. Write down what you expected, what actually happens, and what you have ruled out. Then ask for help or walk away. Writing the summary alone solves an embarrassing percentage of bugs, which is why programmers keep rubber ducks. Running work in Pomodoro-length sessions gives you natural checkpoints where the trance gets broken automatically.

Projects: your brain is not a backlog

Group projects and final-year projects fail the same way: everything lives in someone's head, and that someone is on holiday. Break the project into tasks small enough to finish in a sitting, give every task a home where the whole team can see it, and do a five-minute brain dump whenever the mental tab count gets high. A student planner that holds tasks, deadlines and a calendar in one place is the difference between a project plan and a group chat archaeology project.