How to do research and write research papers — a meta-skill that powers every other domain in this vault. Learn it once and it applies whether you’re researching AI/ML, robotics, chip design, or something you got curious about last week. Built balanced: the universal research process + writing craft as the spine, with different fields’ and paper types’ conventions called out (so you can write an empirical ML paper, a systems paper, or a survey and hit the right register each time).
Source: distilled from the canon of “how to do research” guidance — The Craft of Research (Booth, Colomb & Williams), Simon Peyton Jones’ “How to Write a Great Research Paper,” the “three-pass” paper-reading method (Keshav), Writing for Computer Science (Zobel), and standard scientific-method / peer-review practice. [reference] — a skill you learn by doing (reading real papers, writing a real one), not just reading about.
Every note opens with a plain-language “kid version” before going deep, and ends with a Key insight. Grouped into three parts.
Part A — Doing the Research
- What Research Actually Is — [Beginner] — the mindset, what a “contribution” means, the kinds of research (empirical / theoretical / systems / qualitative)
- Reading Papers — [Beginner] — the three-pass method (fast triage → grasp → deep), reading critically, mapping a field
- Finding a Problem & the Literature Review — [Intermediate] — spotting a real gap, doing a literature review, reference management
- Research Questions & Hypotheses — [Intermediate] ⏳ — framing a good, scoped, answerable question; hypotheses and falsifiability
- Methodology & Experiment Design — [Advanced] ⏳ — sound design, baselines, ablations, controls, validity threats, reproducibility
- Analyzing & Interpreting Results — [Intermediate] ⏳ — honest analysis, statistical significance, negative results, over-claiming
- Research Ethics & Integrity — [Intermediate] ⏳ — plagiarism, p-hacking, authorship, data ethics, open science
Part B — Writing the Paper
- The Anatomy of a Paper — [Intermediate] ⏳ — section by section (abstract → conclusion) and the argument underneath it
- Scientific Writing Craft — [Intermediate] ⏳ — clarity, precision, the story/claim, the common mistakes
- Figures, Tables & Presenting Data — [Intermediate] ⏳ — figures that carry the argument, honest visualization
- Paper Types & Writing Styles — [Intermediate] ⏳ — the different-styles note: empirical ML vs systems vs theory vs robotics/hardware vs survey vs position; CS vs engineering vs science conventions
- Citations, Referencing & Tools — [Beginner] ⏳ — citation styles, BibTeX/Zotero, LaTeX/Overleaf, avoiding plagiarism
Part C — Getting It Out
- Venues & the Publication Process — [Intermediate] ⏳ — arXiv/preprints, conferences vs journals, CS conference culture, choosing where to submit
- Peer Review & Rebuttals — [Advanced] ⏳ — how review works, reading your reviews, writing rebuttals, being a reviewer
- Presenting & Defending — [Intermediate] ⏳ — talks, posters, and defending your work
(⏳ = outline being filled in; the rest are full deep notes.)
How to actually learn this
Reading this course won’t make you a researcher — doing will. The path that works: read Part A, then pick a tiny question and actually investigate it (even a small reproduction of a paper you like); read Part B as you write it up; read Part C when you’re ready to share it (a preprint or a workshop paper is a great first target). One real, finished, imperfect paper teaches more than ten notes.
Related
- Experimentation & A/B Testing — experimental design & statistics, the empirical backbone
- Inferential Statistics — significance, confidence, the stats behind honest results
- MLOps — reproducibility & experiment tracking for ML research
- DSA · Distributed Systems — example fields you might research in
Software engineering
Infrastructure & security
Computer science
Hardware & engineering