Data science is without a doubt the most in-demand field today. No wonder, data scientists with proficient skills are handsomely rewarded in jobs across the world.
Yet, having a side hustle that brings passive income is always a nice feeling. For some, it helps to clear debts while for others it’s an opportunity to unleash their creativity and earn more money in what could be a potential empire of the future.
Now, you could be a data scientist who’s comfortable in the current job or an aspiring one looking to make inroads into data science. Regardless, the ideas I’m about to suggest would certainly help you upskill, earn a good side income as a data scientist, and most importantly be your own boss.
Monetize Datasets And Expose Your Models Through APIs
Despite the hype and demand around data science, it’s no mystery that all the heavy lifting is done by the bigger firms. This fact, at times, can make a budding data scientist feel slightly overwhelmed when starting out on their own.
So, putting out a super cool AI-powered product that competes with the best in business might not be the best of ideas. The big corporations can easily wipe out emerging startups that seem like a threat to their own monopoly.
Ultimately, you can clean up the data and pitch it to your target customers. Alternatively, build a machine learning model out of your solution and deploy it using an easy to API. Platforms like RapidAPI offer a good marketplace to distribute and monetize APIs. A lot of developers with no machine learning expertise look forward to such APIs to plug and play in their apps.
Even better, try collaborating with mobile app developers, and build ML-powered apps. I know, I know, with the advancements in Core ML and TF Lite, there’s on-device inference and re-training possible today. Yet, there are a lot of use cases that the Mobile ML ecosystem hasn’t tapped into or optimized currently.
For instance, a StyleGAN model today takes huge space when shipped within an app. So, you can look to offer custom cloud-based solutions and expose them through APIs for apps. Any fun, creative, out of the box image transformation model has a great chance of being adopted widely.
Despite the saturation of the app marketplace over the years, machine learning-powered AI apps have opened a void that you can tap into and generate recurring revenue from Google and Apple’s App Store.
Help Small Businesses By Doing Consulting Work
Freelancing is a very desirable option and it does provide a viable source of income. Websites like Upwork and freelancer.com are flooded with opportunities for data scientists. But then, to land a good client who also happens to pay well can take a while. Besides, often when it comes to data, a lot of companies are skeptical of hiring a contractor due to privy issues and would rather opt for someone who can work full-time on-site.
So, one of the easiest ways to get started with consulting work is by helping out folks and businesses around you. There are lots of small businesses that possess tons of data yet are miles away from data science and machine learning.
Though, it’s common for local businesses around you to get intimidated by buzz words such as machine learning and artificial intelligence. A lot of them have the feeling that AI could potentially automate or kill their jobs. But then, it’s your role to debunk such misconceptions and help develop strategies for their data.
Start by asking local small or medium businesses if they have any data needs. Then, offer solutions on how your skills can help them in better decision making with respect to data. It could be something as simple as predicting the demand for their goods and forecasting future sales. Or maybe offer to build product recommenders and fraud detection systems for them.
Perhaps, this side hustle might not be as rewarding from a monetary perspective, but it can surely help you grow a network and build partnerships.
Start Blogging To Teach, Learn As Well As Earn
Blogging is one of the best ways of utilizing and honing your skills as a data scientist. In fact, not just data scientists, even developers should share their knowledge through articles, podcasts, or videos.
Now, making a fortune out of blogging doesn’t happen overnight for everybody and certainly requires putting in significant time, effort, and perseverance. But still, you can start off right today, be it through sites like Medium and YouTube, or by hosting your own website or selling video courses (Udemy is a good launchpad) or e-books. For monetizing personal blogs, one can opt for ad revenue or look for sponsorship through platforms like Patreon.
In my opinion, every data scientist should blog as it helps document your work, reach a broader audience, stay updated with trends, establish credibility to get decent projects, and generate a sustainable source of income in the long run.
Try Algo Trading. But Not By Yourself.
Stock market analysis, cryptocurrency predictions, real estate, sports betting are all interconnected with data science today.
So, it’s natural for a data scientist to look up to the world of finance in a bid to leverage their skills for earning money. And it seems like an interesting way of making money as well since your code has a real-time impact on the revenue.
But I’d like to state that algo trading is the riskiest way of earning money. In fact, to put it bluntly, most data scientists actually end up making no money and only burn their capital.
First, besides data science, algorithmic trading requires a great deal of financial knowledge and a solid idea of how the stock market works. Despite the presence of technical trend indicators such as RSI and MACD that can be leveraged in Python scripts, one still needs to develop multiple strategies and test them out on dozens of parameters.
Even if a data scientist manages to pull this off and gets a decent accuracy, still playing with real data is a different ball game altogether. All the major hedge fund firms today already use their own algo trading bots(based on hundreds of parameters and well-tested strategies) in day to day trading. The hedge fund managers can very quickly decipher algorithmic trading patterns and break it as soon as you invest in a good sum of money. Worse, they could inverse their strategies to ensure you incur huge losses.
So, I certainly wouldn’t recommend doing algorithmic trading using real money on your own. Instead, if you happen to devise a trading algorithm that gives proven results on back-testing, pitching it to a broker is always a safer bet to make money.
There are Kaggle competitions as well and they do give out huge rewards to the winners. But I consciously didn’t talk at length on it, as the chances of making money are negligible. Nevertheless, Kaggle does provide a good learning curve while also helping grow your network.
I hope the numerous ideas mentioned above inspire you to start a side hustle as a data scientist. A side hustle is the first stone in a potential business empire and helps keep the entrepreneurial spirit alive.
That’s it for this one. Thanks for reading.