As one of the most significant breakthroughs of the 21st century, it is no wonder that several students want to get involved in artificial intelligence (AI) based jobs. Statistics show that the number of postings per million is growing at a much faster rate than the number of job searches, showing there is currently a lack of talent. However, it is not quite as easy as it sounds. As the number of applications for AI expands into new areas, students simply are not learning the necessary skills because they do not exist yet. Classes at universities are gradually developing but employers do not have the experience to know exactly what they are looking for.
In this fast-moving AI environment, enterprises need talent. The demand has triggered a rise in salary by 11% between 2017 and 2018 with AI engineers being amongst the highest paid company categories. Students need to ensure that they are at the top of the list when companies are seeking talent, and there are various ways to ensure that happens.
Step 1 - Getting started in AI
The field of AI is broad and different skillsets tend to be in demand across diverse industries. However, hiring managers will usually expect candidates to hold a Bachelors degree in Mathematics and Computer Technology as a minimum requirement. These standard degrees will most likely help you get into an entry-level position. If you are looking to become a specialist, universities now offer bespoke qualifications in Data Science, Computer Science and other related disciplines. These courses will normally cover far more advanced theories such as cognitive science, robotics, engineering and physics that go beyond the entry-level degrees.
Whilst at university, it is important that you take the right course to develop your skills. In general, there are four fundamental skillsets that all employers will look for.
- Software engineering – working with programming tools such as Python and TensorFlow to manage large streams of data and make it into a production level model
- Knowledge of machine learning theory – to work out the best models to apply to a given situation
- Statistical inference – the ability to quickly evaluate if the models you build are working
- Domain level knowledge and insight communication – ensure that everybody can understand AI models and drive business solutions. There is little value if you have amazing technical ability but cannot show anyone how to use it.
Step 2 –Picking a specialist application
There are many AI jobs depending on your area of expertise. Each of them requires varying degrees of education and skills so it is a matter of working out which best suits your background. Some of the most common jobs today are:
1. A data scientist works with numbers, analysing large datasets and modelling them. It lies at the heart of AI. Somebody in this role will be expected to have a mathematics-based qualification and advanced knowledge of machine learning.
2. Machine learning engineer is somewhat like a computer programmer. They build an application to perform specific tasks using complex predictive analytics models and large datasets. They will have software development experience and most likely a doctorate in computer science
3. Business intelligence developer focuses on turning data into actionable insights for business. They are more interested in designing and being creative with models albeit with a keen focus on math and computer science
4. Big data engineer/architect focuses on the big data architecture of business with knowledge of technology such as Hadoop. They will understand how data should be stored and processed with more of an IT background than some of the other roles
5. Research/robotics scientist aims to build and maintain robots that do tasks, which imitate human actions. They will have knowledge of engineering and physics to support what they do as well as programming. The role helps deploy the models of data scientists and engineers.
Step 3 – Do not stop learning
Whilst university courses can teach all the technical theory and make you an expert, AI is changing daily and it is vital that you continue learning and researching to keep up with any trend. New courses, many of which are free, pop up on well-known resources and it is a brilliant idea to do as many of these as possible.
Step 4 – Finding a job
As somebody who has studied AI, you will most likely be aware that the way recruiters find talent has changed in recent years. Recruiters will often use AI-based technology to complete the first review of CV's and see who has the right skills for their position. This means you can be removed from consideration by an algorithm without ever speaking to a human. With that in mind, keeping your CV updated across talent sites is essential for getting found. The same applies to social media such as LinkedIn or anything else you use that is work-related. Recruitment algorithms will typically use keywords to find the candidates they need. Ensure all your areas of interests are accounted for as well as their synonyms if possible.
Step 5 – The AI Interview
Employers are keen to see that you understand how AI can be applied to their industry and have a clear vision of how you can help them to achieve it. Depending on the job role, they will ask technical questions about the platforms you develop with, whether that be Python, R or TensorFlow. It is not uncommon for candidates to be presented with complex mathematical problems to solve on the spot.
(The author is a global executive director – AI and big data, Xebia)