There has never been a better time to be a developer. We have more jobs open than we have qualified candidates to be able to fill. This means that if you want to code, to build software, to change the world through tech, there has never been a better time to jump in.
AI, or artificial intelligence, is a term that is increasingly used to cover a broad range of topics. In this first article of this series, I want to be sure you understand the scope of what I am interested in sharing with you.
I have been working in the AI space for the last 5 years or so. I have been helping companies with cloud transformation for decades before that. I know some of the problems the biggest customers face when it comes to leveraging emerging tech. I also know that hearing “the next latest thing” is not always comforting, especially to the person leading the change management system in the organization.
However, this change, AI, is one that makes you better, makes software better, makes companies better. It reduces the amount of code and business logic in our apps and leverages the power of machine learning to improve the decisions our code has to make. It allows us to solve harder problems with less time, and with less code.
One of my favorite pre-built machine learning model genres is speech. Speech services allow us to start thinking in terms of natural human interaction instead of buttons, menus and dropdowns. Today as a developer you can add voice to a company’s brand, not only by using a platform like Alexa Skills, but also create custom speech services embedded into their web or mobile applications. More on this in future articles.
One of the most exciting projects I worked on this year I created a new Unity project for building an AR (Augmented Reality) game. In addition to be able to flick things around and move things with my finger on the screen, I also wanted the players to be able to interact with the game with their voice. I am a developer, I don’t want to spend my time building an algorithm. training it with data, testing, failing and repeating until I have a model that works well. If it is possible, I would much rather use an existing model that solves my problem.
We have a collection of labs for creating mixed reality apps, if you want to build something similar. Look at the 300 level tutorials for infusing AI.
This idea of leveraging pre-built models helps to democratize AI, allowing anyone to build apps that infuse AI models without having to build the models from scratch. Now I can add speech, text analysis, search, and computer vision models to help me solve a problem.
I encourage you to learn as much as you can about the pre-built models that are available. This will help you make sure that you only use your data science team for things that cannot be easily solved with something that already exists. Make sure the right teams are building the right stuff. Do not overlap with work already being done by companies, like Microsoft, that specialize in this. As more models are commoditized, your applications will quickly fall behind if your developers (or you as a developer) do not start infusing AI into them TODAY.
Considering that companies like Microsoft have been creating these models, for decades in some cases, it only makes sense that we, as developers, would leverage those models as opposed to attempting to build one ourselves, from scratch.
Stand on the shoulder of giants and build incredible tech, but be careful not to reinvent something that already exists and partner with your vendor early to ensure you have the right support for your project and use case.
If you are a developer just getting into this space, review each pre-built model. Microsoft makes it easy by letting you test models right in the browser. You can find these demos here:
Examine the metadata you get back from each service. Think about how you could use these services to improve your existing applications. Can you use text analysis to determine sentiment of product reviews? can you use facial recognition to track high-rollers in your casino or hotel? Can you use language to create an engaging customer chat bot? Can you use cognitive search to make sense of unstructured data in the cloud?
In this series we will investigate these use cases, and hundreds of others. Have an idea for us to feature? Please email me at Noelle @ AILeadershipInstitute.com
Thanks for reading, subscribing and staying connected!
P.S. If you are a business leader and interested in AI, check out the AI Leadership Institute!