Volta Venture

What should be considered when selecting a Data Science Unicorn?

Alan Hylands, in his article "Generalists Are The Real Data Science Unicorn," has taken a closer look at this problem and described his view.

Data Engineering. Data Analysis. Data Transformation. Machine Learning. Artificial Intelligence etc...

If we look at any job posting, a Data Science Unicorn must be proficient in all areas of data science to be truly interesting. They also need to have a Ph.D., maybe even a couple of master's degrees, and lots of side projects on Github. Not to forget, of course, the relevant internships and work experience. And most of the time, the list does not end there.

However, it's worth asking yourself one important question: Is it worth looking for a specialist in a specific area, such as machine learning or data engineering, or finding someone who has general knowledge—do they approach the topics broadly and superficially, but have the germ of knowledge on each topic?

Finding a specialist seems like a promising option, especially when it comes to working for a large company. The problem for Data Science Unicorns, however, might be that early in their careers it's difficult to learn about different areas of data science to really know which direction to focus on.

But what if you are running a smaller company, such as a small business or early-stage startup? With a small data team, you are more likely to be looking for new hires with skills that cover multiple bases at once. This might lead you to choose Option 2—Generalist.

So make sure you build the right infrastructure to get the most out of your data/analytics/data science. A machine learning engineer, regardless of talent and intelligence, is simply not the right person for the job. You need a generalist.

AI is creating enormous wealth and every month will continue to increase it. So why shouldn't you and your company take advantage of this opportunity and start implementing AI in your business? It's time to democratize access to AI!

Of course, AI projects are expensive to build- they require highly skilled engineers, and it can cost millions of dollars to develop a one-size-fits-all AI system, such as one that improves web search. For this reason, only large tech companies can afford such innovations and make huge profits from them. But what about smaller companies and individuals for whom one-size-fits-all AI systems do not work due to very specific and individual data?

Ng explained in his TED talk, "How AI Could Empower Any Business," that we are getting closer every day to a widely available AI system. And we don't have to be experts in coding! There are new AI development platforms that shift the focus from asking people to write a lot of code to providing data. This way any small business can use these platforms to make their work easier or increase revenue by leveraging their specific data.

Andrew shared an amazing example of a platform his team developed that helps detect defects in fabric. You can easily show the AI what tears or discoloration look like in the fabric by drawing rectangles on provided images - helping the AI get smarter and eventually detect such defects itself.

" Building AI systems has been out of reach for most people, but that does not have to be the case. In the coming era for AI, we'll empower everyone to build AI systems for themselves, and I think that will be incredibly exciting future." Andrew Ng.

Listen to Andrew Ng TED talk here:
https://pos.li/2mgq6y

How AI Could Empower Any Business | Andrew Ng | TED Expensive to build and often needing highly skilled engineers to maintain, artificial intelligence systems generally only pay off for large tech companies wi...