Blog: How to Become a Freelance Data Scientist
When the demand for workers in a particular professional field rises, it tends to be easier to act as a freelancer in that profession. Routinely, companies may prefer to hire full-time staff, but if they’re scrambling to fill roles, hiring an independent contractor may be on the table.
We’re seeing this juxtaposition today in the field of data science — more data scientists are needed than are available to fill open positions. This situation presents a distinct opportunity for data scientists who’d prefer, for a variety of reasons, to operate as a freelancer rather than a full-time employee. I am personally taking advantage of this market dynamic and I’m loving it! In this article, I shall review the tried and proven methods I’ve used to flourish as a data science freelancer.
Achieve Guru Status
As a freelancer, it’s very important to know your stuff technically. There’s no such thing as a trainee or novice freelancer. You’re expected to be an expert in the field. This means that before you attempt to go the freelance route, you must do your homework and seriously tool-up for the gig. The people hiring you will necessarily qualify your abilities, so be prepared to answer tough questions. This means you need to have a robust and relevant academic background and years of practical experience with employment references readily available. This status could mean years of hard work to achieve, but it will be well worth the trouble.
Build Your Brand
Freelancing is like running a business of one employee, and like any business you need to be able to market the business — namely yourself. You need to build your brand. I’ve used several methods that have worked well for me. First, you guessed it, I write a lot of articles on data science and machine learning. Writing articles isn’t necessarily easy to do. You must actually know a lot about the subject, and you also need to be able to articulate it well and have well thought out opinions and views that you can effectively justify. The ability to write critically about a technical field takes time to really understand the industry and its players.
Another way to build your brand is to speak at conferences. Speaking establishes you as an expert, and believe it or not, you actually learn a lot by speaking since you tend to think seriously and research ideas when preparing your presentation. Plus, audience questions may lead you to new perspectives you hadn’t considered before.
You can also teach classes on data science-related topics. There are many, many teaching opportunities these days, so finding a traditional school, boot camp provider, or MOOC may be just the way to get some street cred by helping others. Again, you learn a lot by teaching a subject, especially data science.
Although very time consuming, you can try writing a book on the subject. There’s nothing like showing off your book during that first meeting you have with a new freelance client.
To become a successful freelancer you also need to let potential clients know you’re out there and available. Fortunately, social media is a great (and free) way to manage this. Being active on Twitter is useful especially if you’re able to get recognizable names (people and companies) to follow you. Frequenting data science-oriented LinkedIn groups is also a good strategy (by making comments and writing short articles). Answering questions on Stack Overflow and Quora lets you become a recognizable source of good information.
Attending or even organizing local Meetup groups on data science related subjects is a good way to network. Once again, making short presentations gives you a platform to shine and potentially attracts clients and/or strategic partners.
There are many ways to become a data science freelancer these days, you just need to be creative. It takes time and effort that’s typically not reimbursed well up front, but the methods I’ve outlined above are all a form of technical marketing that will be revenue producing in the end. I’ve even gone back to my alma mater UCLA to present at career day events, as well as mentor undergrad and graduate students interested in getting into the field. I believe that “giving back” and helping others succeed sort of completes the cycle of being a data science freelancer. Best of luck!