Developing more than five successful apps is a feat for anyone. But when each app is built at the intersection of machine learning and mobile engineering, it’s particularly impressive. Meet Melody Yang: the genius and founder behind the Japanese language learning app Nukon, the Instagram posting time app PopTag, the mentorship app Relate, the image classifier app Recognize, recipe sharing app Share Recipe, and the weather-based outfit recommendation app WePick.
These apps all have two main features in common: they exhibit machine learning, and were developed using mobile engineering. In a world that’s increasingly leaning into artificial intelligence, machine learning – a subset of artificial intelligence – is becoming a must-learn skill. Machine learning refers to computer algorithms that improve and learn on their own based on the input they receive. This is revolutionizing tech products, by helping tech get to know users and their habits better.
Accordingly, machine learning is being incorporated into many daily technologies. How your social media feed is populated is dependent upon machine learning. When your maps app in your phone suggestions directions to a spot that you frequent, that is machine learning. And, consumers are becoming more and more reliant on this type of technological help. Many tech companies are now pushing the limits: how much can tech help us in our day-to-day lives? How can we incorporate machine learning in apps, so that an algorithm that understands us is readily accessible?
This is where upcoming college graduates or hopeful job seekers should take notes. As the demand for artificial intelligence intensifies, the need for highly qualified mobile engineers or those who are proficient in machine learning will increase, too.
The Rarity of an Overlap Between Machine Learning and Mobile Engineering
Neither machine learning nor mobile engineering are easily learned or grasped. Both require extensive studying and experience within their fields, which is what makes Yang’s overlapping knowledge and understanding of both so rare.
“Since there’s only a little overlapping (ie. coding) between skillsets for mobile development and machine learning, people tend to choose one route and stick with it. If one is heavily interested in consumer-facing products, they may find doing research tedious and vice versa,” Yang explained. “Instead of learning too many things and being a master of none, being really good at one makes candidates more hirable. However, by mastering both, one becomes invaluable because they can turn research into real-world applications that solve hard problems. It takes time, efforts, and motivation to become such an individual.”
As job seekers are looking to increase their hire-ability, many wonder if they should look beyond developing just one core skillset to make themselves more versatile. Yang says this isn’t readily necessary. “When I look at positions in the job market, each of them is designed for candidates who are masters at one specific skill set. It’s rare to come across a job posting that asks for skills in two distinct domains.”
So, why have both? It works for Yang as a founder, who can build mobile apps that incorporate machine learning, keeping everything in house.. “By mastering both, one becomes invaluable because they can turn research into real-world applications that solve hard problems,” she commented.
Enter: the Cloud for constant machine learning updates
Tech relies on updates – even machine learning tech, that can technically “update” itself. Yet, Yang made the bold choice to build her machine learning algorithms ‘locally’ as opposed to in the ‘cloud’. The ‘cloud’ refers to the ability to access or update software at any time, in the same way we can upload photos to the cloud or download files from the cloud via our phones or laptops. Without being stored on the hard-drive, it can still be accessible, and is always connected. On the other hand, when the code is ‘local,’ that means it isn’t able to be updated from the cloud, because it’s now native to the code that exists on the device (or, in this case, the app’s code that is downloadable from the App store).
Yang chose this avenue carefully, knowing its risks because she also recognized the risk of how cloud-based machine learning can go wrong. “While cloud-based comes with the flexibility to update machine learning models anytime, the need for network access is its downside. Having machine learning implementations locally, namely, on-device machine learning, has better performance as it’s not relying on network requests,” she commented.
“Taking my app Nukon that helps people learn Japanese language with machine learning as an example, most users build their habits of frequenting the app. It’s frustrating when the app can’t give you feedback on your learning simply because your network is out and the machine learning feature can’t function properly. Hence the users’ learning experience gets disrupted. It becomes hard for users to maintain their habits of learning. My priority has been providing the best user experience possible,” Yang explained.
With a focus on user experience and machine learning, Yang is applying her hard-earned skill sets to many industries and solving many types of problems. After all, if you have experience in both machine learning and mobile engineering, there isn’t much you can’t build.