Technical Writing

Introduction
Next month (September 2021) marks my two-year anniversary of being active on Medium. Writing, less so technical writing, wasn’t my thing. It started as a distraction from working on a book-length dissertation but ended up as a two-year-long commitment with many more to come.
I’ve put up so much effort to make it work and created 50+ original blog posts on various domains of Data Science, including Programming (R, Python, and SQL), Machine Learning (supervised and unsupervised), and Experimentation and Causal Inference (A/B tests, quasi-experimental designs, and observational designs).
In return, I’ve benefited tremendously from my writing, a loyal readership, a close community, and the extra writer bonus three times in a row (can’t complain), just to name a few.
In today’s post, let’s talk about the reasons why you should also start a technical blog and how.
1. Create an online portfolio that showcases my skillsets
My academic background is both a blessing and a curse. Completing a competitive Ph.D. program showcases my intellectual ability to learn things fast. But, in the meanwhile, companies may wonder:
"How much practical experience do academics have?"
Trust me. You have no idea how many times I got asked this question.
From the company’s side, it represents a healthy dosage of hesitation or a reasonable amount of concern as the stake on the table is high.
Updating an online technical blog sends a strong signal. That is, I know the subject well enough and decide to go the extra mile: writing a step-by-step tutorial on it. There are mainly three career tracks within Data Science: Analytics, Inference, and Algorithm. Pick the niche that best fits your background and tailor the content and writing styles to the long-term career trajectory, which sends an even stronger signal.
Compared to a personal website or other alternatives, a technical blog is a more engaging form of representation that facilitates two-way conversation, which is lacking in a website. Moreover, Medium offers a ton of cool functionalities like note-taking, highlights, and sharing that make things much easier for your audiences to engage with the content.
2. Better understand the material
Teaching is the best way of learning.
I used to think statistics concepts are easy but can’t explain p-value and hypothesis testing without using any terms. I used to believe I understand Experimentation and Causal Inference until I got stuck with a technical detail of Matching in an interview.
There are so many other "I thought but not in reality" situations. We maintain a short memory of concepts and methods learned from a textbook but forget them quickly if we do not refresh our memory.
From the pedagogical perspective, technical writing is the process of delivering what you have learned. Students who go through the process can better understand the content. I’m always humbled by how little I know about the subject until taking notes or writing something.
3. Connect with the community
Like-minded folks keep the same habit.
As an avid reader, I spend hours reading others’ work on and beyond Medium. It has been a long journey, and I’ve made so many meaningful connections along the way.
Fellow Data Scientists reach out for potential collaboration opportunities.
PMs are curious about the business application.
VCs who are investing heavily in the Experimentation and Causal Inference domain would like to know its future.
By creating content that benefits others, I’ve grown my professional network significantly and connected with so many other interesting folks. Here are some numbers:
~1,700 followers on Medium
~ 3,000 followers on LinkedIn and counting
One of the best suggestions that I’ve received is your writing is your brand. By creating a brand, you will receive traction in the industry.
4. Help others better understand the material
There are a ton of online courses that try to sell the idea that "Learn Data Science in 10 weeks or less," "Cracking Data Science Interviews and earn you first million."
This type of clickbait content is ubiquitous on the web, and beginners may be lost in the way of searching for the real stuff. So, my biggest writing motivation is to democratize Data Science by creating authentic content that is accessible to everyone.
My writing strategy is to do a thorough "literature review" of the existing works, spot the incremental value (i.e., the gap), and write a series of original posts on the topic.

Where to Start
Medium is the best writing platform that supports content writing. I enjoy reading and writing on the platform. It offers a flexible writing style, and its note-taking function is the best in the industry.
If you are a fresh starter on Data Science, try to write on topics like fundamental concepts, such as what is cross-validation in Machine Learning, what bootstrap is in Statistics, and other basic concepts.
If you are more experienced with these topics, try to take up more challenging topics, like writing on a hands-on tutorial, e.g., how to build a decision tree in Python.
Start with the most comfortable level with you and iterate quickly to the next level, which is the key to successful technical writing. My first blog post on the platform is unreadable, to say the least. However, my writing style has improved significantly over the past two years. Start small and iterate quickly.
Thanks for reading so far! If I have inspired you to write something, please let me know in the comment. I’ll give your first post a push on my social media (e.g., LinkedIn and Twitter).
Takeaways
1. Create an online portfolio that showcases my skillsets.
2. Understand the material better.
3. Connect with the community.
4. Help others better understand the material.
_If you find my post useful and want to learn more about my other content, plz check out the entire content repository here: https://linktr.ee/leihua_ye._
Enjoy reading this one?
Please find me on LinkedIn and Twitter.
Also, check my other posts on Artificial Intelligence and Machine Learning.