The changing face of machine learning

Whether it’s offering YouTube recommendations, developing self-driving cars or solving the mysteries of the building blocks of life, uses for machine learning or artificial intelligence are popping up everywhere. So it’s no surprise that LinkedIn has identified it as one of the jobs on the rise for 2022.

If it sounds intimidating, you’re not the only one who feels that way. Lila Ibrahim, chief operating officer of DeepMind, an offshoot of Google, told the Financial Times that even she suffered from “imposter syndrome”working around the company’s “super-smart people”. But it’s a big and growing industry and the entry requirements aren’t always that steep.

What is the job like?

As a machine learning engineer, you teach a computer how to learn to perform tasks without being directed by a human. It means writing code and algorithms while managing huge amounts of data, then running tests on what you’ve created.

That means you’ll need top-notch maths skills, an understanding of programming languages like R and Python and an analytical mind. Because you’re often working on big problems, you’ll be part of a bigger team, so communication skills are also key, says one former engineer.

At the highest level, you could be making serious scientific breakthroughs. Ibrahim says DeepMind engineers are driven by the “mission” of trying to build artificial intelligence and working on inspiring projects – such as one that allows motor neurone disease patients to preserve their voices digitally as their illness progresses.

What qualifications do you need?

You’ll usually need a relevant degree or even a postgraduate qualification, the National Careers Service says. Maths, statistics, computer science and data science degrees are all likely to be helpful.

There are also degree apprenticeships and higher apprenticeships, where you can get the equivalent of an honours degree or a master’s while you earn and gain experience at work. You’ll usually need four or five GCSEs at grades 9 to 4 (A* to C) plus A levels or equivalents.

At the highest end of the job market, you’ll be expected to have a PhD. But DeepMind’s engineering lead Chris Gamble told Business Insider that a degree in AI or a computer-science related field isn’t always necessary to get a job at the company. “We look for the potential to succeed, a demonstrated ability in AI, an ability to code, and an entrepreneurial drive,” he said.

What is the career path like?

You could start in a more general role in data science before specialising in machine learning or artificial intelligence, the National Careers Service suggests. There may also be graduate schemes allowing you to get a start in data science at government departments, the NHS, universities, or companies in finance, IT or sales.

Once you’re working as a machine learning engineer, you could take on responsibility for more people or be a project manager. You could move between industry sectors or go into academic research and teaching. Or you could become a freelance consultant.

How much can you get paid?

New starters can expect a salary of around £30,000 as a data scientist, the National Careers Service suggests. With experience, that can rise to £70,000.

Meanwhile, at the top of the profession, the Financial Times says that senior scientists at DeepMind can be paid around £500,000