By AbdulMajedRaja, Analyst at Cisco
Data Science is one the of the few domains where every college graduate wants to get in. It’s primarily driven by the amount of hype this domain has received in the past — starting from HBR’s “Data Scientist: The Sexiest Job of the 21st Century” article to the recent developer surveys that put Data Scientist as one of the most paid roles. Riding on this hype, Training Institutes, Educational Institutes and Bootcamps started introducing Business Analytics / Machine Learning / Data Science / Artificial Intelligence courses which a lot of working professionals and fresh grads take with the hope that these courses would land them in a Data Science Job. But it hardly does!
Because, any sane recruiter hardly cares about the names of Courses one has taken rather the emphasis is actually on “The Portfolio” — The stuff that you have done/built to show your passion for Data / Numbers. It could be anything from a Visualization of Public Data to a Github Code of your project with a well-read documentation.
“If you don’t produce, you won’t thrive — no matter how skilled or talented you are.” — Cal Newport
Before we dive in further, I’d like to set a framework for what could make a compelling Portfolio. The portfolio that you would want to build, should be capable of doing the following to you:
- More findable on Google (Search)
- Create a Branding
- Allow building Fan Base / Audience
- It’s something that you can offer your potential employer to evaluate your expertise.
- Ultimately, A good reputation in the domain that you want to grow further
The Point — Kaggle Kernels
In this article, I’m trying to make a point of how one can show off their Data science skills with Kaggle Kernels — where you can build your portfolio — which could be either Visualizations with Storytelling or the state-of-art Neural Nets Implementations.
The Kaggle Myth — Competitions Track
The name Kaggle highly resonates with the data science community for a Competitive Machine Learning Platform — much like Topcoder/HackerRank — in the computer science community. Because of that, a lot of beginners fear entering the world of Kaggle, with this naive assumption that Kaggle is a place for Pros. What most of them (including me at one point) forgot is that, those Pros (who are Grand Masters or Masters in Kaggle Ranking) once were beginners when they joined Kaggle. Moreover, Competitions are not just what Kaggle is all about.
Competitions are just one of the tracks available on Kaggle. Kernels is another very powerful track on Kaggle. Kaggle Kernels are very good resource to learn something and also to share something — thus, making it use to build your Data Science Portfolio.
Advantages of Kaggle Kernels — as your Portfolio
- Kaggle Kernels are indexed by Google which means, If someone on the planet is looking for something relevant to your Kernel, it’s going to be displayed as a Google Search result
- Kaggle Kernels — votes / medals contribute to Kernel Progression System and Ranking. i.e., You can entirely stay from Competitions track, and still be an Expert/Master/GrandMaster — just by focusing on Kaggle Kernels
- Awards/Fame/Money — Anytime you win something on a platform where you have global audience, you’re getting a global fame. That stays true for Kaggle too. Kaggle often hosts Kernel-based competitions with Cash Prizes or Kaggle Swags. The latest is Kernel Author of the Month with a cash prize (upto $2000) associated with it!
- Audience/Fan Base — Kaggle Profile comes with Followers/Following option. That means, any Kaggler interested in your work can also follow you on Kaggle. It also lets you add other professional profiles like Linkedin, Github where people can follow you further beyond Kaggle.
- Kaggle is well-known — Kaggle is very well-known among sane recruiters who are familiar with Data Science/Analytics/Machine Learning. That means, adding a decent enough Kaggle profile link in your resume is going to give a lot of extra edge and you’re riding on the fame that Kaggle has built for it in the Data Science Market.
While it’s advisable to have a diversified portfolio — let’s say blog posts, Github Codes, Slideshare presentations. For those who can’t manage the multi-faceted portfolio, Kaggle Kernels platform serves as a powerful alternative for all these. While fitting in the portfolio framework that we discussed above, Kaggle Kernels can also etch your name among the Data Science community.
Thus, you can simply show off your data science skills with Kaggle Kernels which ultimately could help you land in a job.
Let me know your thoughts in the comments, even if you agree / disagree with me!
- Kaggle Kernels
- Kaggle Kernel Author of the Month
- Kaggle SRK Profile
- Announcing fast.ai part 1 now available as Kaggle Kernels
Bio: AbdulMajedRaja is an Analyst at Cisco.
Original. Reposted with permission.
- To Kaggle Or Not
- How to Build a Data Science Portfolio
- Building a Data Science Portfolio: Machine Learning Project Part 1