DIY curriculum for post-school life

post-school

Last February, I defended my PhD thesis and graduated from more than 2 decades of school life. Now, it’s been a full year of post-school life. There are no more exams and curriculum to quantify my GPA. In this post-school life, I start to realize that I have to be the one setting my own goal, designing my own curriculum, and evaluating my progress introspectively. In this post, I am sharing some lessons that I find useful in DIY curriculum. 

Develop soft skills in the hard way

Seek mentorship and be a good listener

In post-school life, the first challenge is that there are no more designated teachers whose primary job is to teach people. Everyone has their own job to work on and their own problems to solve. It becomes critical to find someone who is willing to take the time and share their wisdom. 

Confucius said, “When I walk along with two others, they may serve me as my teachers. I will select their good qualities and follow them, their bad qualities and avoid them.” Jordan Peterson in his book 12 Rules for Life said, “Assume that the person you are listening to might know something you don’t.”  

I am fortunate to have great colleagues, friends, and peers who are very glad to answer my questions and engage in open discussions about technical topics and career development. Before seeking mentorship, I would first organize my thoughts and questions concisely on very specific topics, rather than randomly touching wide topics superficially. Then, I would contact potential mentors, such as senior coworkers, peers, friends, and ask if they would have time to meet and discuss my list of questions. With agenda at hand, it is much easier to stay on track and understand how far I am away from my goal. During meetings, the most crucial thing is to listen. Meeting with mentors is usually not a place for me to express my opinions, but to open up and listen to different opinions, advice, and experience. Sometimes, I do engage in heated debate with my mentors, mostly on open-ended issues and brainstorming. 

Mentorship should be mutual. Mentees benefit from the rich experience and knowledge from mentors, but what do mentors get from this interaction? After all, mentors are not designated teachers. I do not have a clear answer to this question yet. From my experience, sharing with others and seeing others grow with my help gives me a huge sense of achievement. 

Set boundaries

The second challenge in post-school life is that there is much more interpersonal interaction than at school. Most projects involve collaboration and coordination, and a great amount of my working hours are occupied by meetings. Other teams may need help, and want to reprioritize my tasks. For example, someone may say, “Can you help me check this? It should be very quick.” It is one thing to be willing to help others, but another thing to be dragged around by others’ request and not have time for my own work. When getting ad-hoc requests, it is important to recognize their urgency and importance, and involve my managers if reprioritization is needed. 

When disagreement happens (which is very common), it is important to defend the ground without being offensive, a lesson I learned the hard way. Focus on the fact, numbers, and results, not opinions. We have to speak up for ourselves and our ideas. It is not a good idea to always say “yes”.

Try and fail fast

When it comes to interpersonal interaction, a lot of times, there are no clear rules and single correct answers. One communication style may work for one person, but not another. The only way to find out is to try it. Sometimes I just have to take risks, speak up, and make mistakes. Making mistakes and learning from them in early career is better than avoiding new opportunities. As a mentor said and I paraphrased, “It’s better to say sorry later than not doing anything at first.”

Sharpen hard skills in the diligent way

Both on and off the job

Ad-tech specific knowledge and expertise, company specific infrastructure, framework and toolings, product oriented research and development… Surrounded by talented and experienced data scientists and engineers, I am constantly learning on the job, about how to apply data science to business problems, and how to incorporate research insights into products. In my first post-school year, I have written and organized personal learning notes on various issues such as Scala, Spark, Airflow, Google Cloud, and documented my discoveries on company internal wikipages on topics such as Feature Engineering and Model Selection. 

Daily job and tasks usually only cover a very deep and specific field, and I find it extremely helpful to broaden my knowledge and aim at becoming a generalist. Starting from reading textbooks on the most fundamental statistical learning and numerical optimization, to following recent publications online, I realize that machine learning and data science is a fast-evolving and dynamic field, and that width and depth are both critical to advance my career. 

Get hands dirty

It’s always been a pleasure for me to read papers, compose a literature review, and come up with a research initiative proposal. However, when I first started to work, I was reluctant to implement an algorithm myself and compose code to deal with real data, and I was incorrectly thinking that implementation was mechanical labor work without any creativity. My misconception was quickly gone once I begun to get my hands dirty in the first implementation task: it was anything but mechanical. Theorizing and research in a controlled condition is neat, but data in the real world is anything but neat. To make an algorithm work at scale with clean and organized code requires as much creativity as doing research, and experience in software engineering. I start to appreciate the ingenious design of a lot of algorithms once I get to implement them myself: knowledge is cheap, show me the code.

Do the work first and worry about “impact” later

I remember when I was preparing for interview questions, one question was “why do you want to transition from academia to data science?” and my answer was “I want to make real-world impact.” I also remember a common tech slogan “We are here to revolutionize the way you do XYZ” (see Silicon Valley TechCrunch Disrupt Parody). “Impact” has become a placeholder and does not convey too much actual meaning. As one of my friends said and I paraphrased, “Your position defines your impact. If you are CEO, any decision you make will be of great impact; if you are an entry level individual contributor, just do your work and worry about ‘impact’ later.” Some tasks at work may not be a million dollar question, and may be solved decently well without sophisticated methods. For example, if you can provide great insights using a simple histogram, then do it. Simple is best. 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.