My process of reading, understanding, and remembering research papers

Watch the video explanation ➔

Reading research papers is a crucial activity for engineers, as it helps elevate their game and deepen their understanding of various subjects. In this article, I will outline the process I follow to discover, read, understand, and remember research papers effectively. Let’s dive right into it!

Step 1: Discovering New Papers

To discover new papers, I rely on three primary sources:

  1. Google Scholar: Google Scholar is an ultimate website for research papers, where you can find a wide range of scholarly articles. I simply search for topics that intrigue me, such as “columnar database,” and explore the top-ranked papers that appear. Google’s ranking algorithm ensures high-quality papers are usually among the top results.

  2. Hacker News: Hacker News is a vibrant community where engineers share interesting findings and ideas. I follow Hacker News and explore papers that are shared by fellow engineers, which often leads me to new and exciting research.

  3. Social Networks: Platforms like LinkedIn serve as additional sources of paper recommendations. When people in my network share research papers they find interesting, I add those papers to my reading list.

It’s worth noting that many individuals and organizations have curated lists of the top papers to read. When starting out, it can be helpful to follow one of these lists without worrying too much about finding the absolute best papers. Following the right people on social networks can also help you become a better engineer.

Step 2: Choosing a Paper to Read

Once I have a collection of papers, I don’t spend too much time contemplating which one to read first. Instead, I encourage picking any title that seems interesting or amusing. Rather than waiting for the “perfect” paper that matches your skills or relevance to your job, it’s better to start reading and broadening your horizons. Remember, reading research papers is meant to be fun and enlightening.

For example, if I come across a top 100 papers list and find a title like “Google File System,” it immediately catches my attention as it sounds fascinating. So, I select that paper and proceed with the reading process.

Step 3: The Three Iterations

To effectively read and understand a research paper, I follow a three-iteration process. Each iteration builds upon the previous one, gradually increasing the depth of comprehension.

Iteration 1: Skimming the Paper

In the first iteration, I skim through the entire paper without trying to understand every detail. Using my favorite PDF reader, I quickly go through each line, forming a rough mental map of the paper’s structure and content. This helps me grasp the overall theme, identify keywords, and establish connections between different sections. While skimming, I selectively highlight the sections that catch my interest, creating a visual guide for deeper exploration. This initial pass typically takes around two hours.

Iteration 2: Thorough Reading

The second iteration is the most intensive. Here, I read every line of the paper meticulously, striving to understand the meaning and its relevance to the larger context. I highlight key details, aggressively capturing information that intrigues me. By making sense of each line, I strengthen the connections between the broader concepts established during the skimming phase. This in-depth reading process can take anywhere from one to two days, depending on the complexity and length of the paper.

Iteration 3: Creating Handwritten Notes

In the third iteration, I delve deeper into the paper, taking meticulous notes as if I were teaching the concepts to someone else. This step is crucial for solidifying my understanding and bridging any remaining gaps. By preparing handwritten notes, I challenge my comprehension and ensure that I have a complete grasp of the material. If there are any unconnected dots or unclear sections, I retrace my steps, revisiting those areas until a strong connection is established.

This final pass ensures a strongly connected graph of knowledge in my mind, enabling me to confidently discuss and explain the paper’s contents. Although not every paper requires this level of thoroughness, for those I truly wish to explore in depth, the third iteration is an indispensable component of my process. It may be demanding, but the rewards in terms of knowledge expansion and expertise enhancement make it well worth the effort.

Conclusion

In conclusion, reading research papers is a valuable habit for engineers to cultivate. By following a systematic process of discovery and comprehension, engineers can effectively navigate through the vast landscape of scientific literature. The three-step approach of discovering papers through sources like Google Scholar, Hacker News, and social networks, followed by three iterations of skimming, thorough reading, and note-taking, ensure a comprehensive understanding of the material. While the process may be challenging and time-consuming, the rewards are worth the effort, as it leads to elevated skills, deeper insights, and a broader perspective in the field of engineering. So, embrace the journey of exploring research papers and let it propel your engineering career to new heights.

Here's the video ⤵

Courses I teach

Alongside my daily work, I also teach some highly practical courses, with a no-fluff no-nonsense approach, that are designed to spark engineering curiosity and help you ace your career.


System Design Masterclass

A no-fluff masterclass that helps experienced engineers form the right intuition to design and implement highly scalable, fault-tolerant, extensible, and available systems.


Details →

System Design for Beginners

An in-depth and self-paced course for absolute beginners to become great at designing and implementing scalable, available, and extensible systems.


Details →

Redis Internals

A self-paced and hands-on course covering Redis internals - data structures, algorithms, and some core features by re-implementing them in Go.


Details →


Writings and Learnings

Blogs

Bookshelf

Papershelf


Arpit's Newsletter read by 80000+ engineers

Weekly essays on real-world system design, distributed systems, or a deep dive into some super-clever algorithm.