World renowned data scientist, researcher, serial entrepreneur and legendary maverick, Jeremy Howard talks exclusively to AIMed. Part one of a fascinating interview..
Jeremy Howard is a data scientist, researcher, developer, educator, and entrepreneur. A founding researcher at fast.ai, a research institute dedicated to making deep learning more accessible, he is also a distinguished research scientist at the University of San Francisco, the chair of WAMRI and is Chief Scientist at platform.ai.
Previously, Jeremy was the founding CEO of Enlitic, the first company to apply deep learning to medicine, and was selected as one of the world’s top 50 smartest companies by MIT Tech Review two years running. He was the President and Chief Scientist of the data science platform Kaggle, where he was the top ranked participant in international machine learning competitions two years running.
His ability to demystify AI and make it accessible to the general public have seen him make many media appearances around the world. Jeremy has invested in, mentored, and advised numerous startups, and contributed to many open source projects. He is also a co-founder of the global Masks4All movement – an all-volunteer organisation that started and powered the movement for people and governments to follow the overwhelming scientific evidence for the need to wear homemade masks in public to slow COVID-19.
You’ve been involved in numerous successful projects, which one has given you the most satisfaction?
Without doubt, helping to set up Masks4All during the pandemic. I did it on a voluntary basis as I was so passionate about it. It started off as a little case study project for my class to teach them about evidence-based data analysis. I quickly realised it was really compelling and important so I got a team of 19 of the world’s top experts to write a paper about it which got published in the Washington Post. From there it grew further and we were soon advising governments around the world. It was a huge project, incredibly successful and saved God knows how many lives.
Does it make you angry when you see people still refusing to wear masks?
I don’t get angry because I understand that people can be misinformed and there’s plenty of misinformation out there. Some people genuinely believe it’s a mark of the devil or a mass conspiracy to create subservience. It’s weird crackpot stuff, but people are entitled to have their own beliefs . Anyway, I retired from that a few months ago because it was too much of a toll on me personally. And it was clear that we had won with the mass take-up of mask-wearing. Once the WHO started supporting us, it got to the point where I realised I wasn’t needed anymore so I went back to my normal life.
What did you want to be as a child?
From the age of seven, I honestly wanted to be a nuclear physicist. People laughed at me and thought I was weird but it all stemmed from a cool little book I read about nuclear energy. It just blew my mind and I knew that’s what I wanted to get into. I kept that ambition right up until I went to university which was when I discovered my maths wasn’t nearly strong enough so I wasn’t able to follow that up. I started studying science and arts and then switched to commerce and then finally a BA with a major in philosophy. But while I was at university I was also working full time at consulting company, McKinsey so I didn’t go to any lectures. I basically studied during two weeks that I took off every semester. It was during that fortnight that I’d figure out what assignments I was meant to have done and then try to find a way to hand them in months late. I was working at McKinsey 80-100 hours a week so it was pretty hectic with very little time for studying.
What sparked your interest in data science?
I remember at school a teacher in year 11 showed us a spreadsheet which I’d never seen before and I thought it was just the coolest thing I’d ever seen and I dived in. Our neighbour at home was a management consultant so I started doing some work for him on a spreadsheet which he loved. Not many people were using spreadsheets then so it was pretty rare. He told me about a new piece of software coming out called Microsoft Access which was pretty much the first personal database building programme so I dived into that.
So by the time I finished high school I was pretty adapt at using Excel and Access. When I got to McKinsey at the age of 18, I quickly realised that everybody there was an expert in stuff – either telecommunications, marketing, agriculture or whatever – and I wasn’t an expert in anything except these computer programmes. But it turned out that everything the consultants were doing had a data analysis component. So I found that all these much older consultants kept asking me for my help with their data analysis. Through that I taught myself operations research, statistics etc so I became a specialist in that. That was 25 years ago and I’ve basically been doing that ever since.
Who’s been the biggest influence on your career?
The internet. I grew up in Australia which is a very long way away from everything and I didn’t really have any peers. So I was on the internet from when I was 17. When I started my first companies, all the people I hired were people I’d never met before, but had worked with through the internet. So this remote work thing that we’re all doing because of COVID is how I’ve worked all my life.
What advice would you give to someone starting out a career in data science?
Be broad not deep. Jobs like management consulting are great, as long as you only stay there two years. Focus on who you’re working with rather than what work you’re doing. One of the most important things in my life has been a decision I made when I first started working as a teenager. And that was to spend at least half of every day learning something new or practicing something new. It’s great because it means I’ve now got all these different skills that I’ve built over time. By practicing new things every day, you soon find that you can pretty much do most things much faster than everybody else around you.
What’s been the biggest disappointment of your career?
It’s more of a regret than a disappointment – and it’s that I stayed in management consulting far too long. I planned to do it for two years to learn about business which I think was a good idea, but I stayed there for eight years, which was far too long. I just wasn’t developing. I became a good management consultant but my strength is data analysis and I wasn’t surrounded by the best of the best. So I wasn’t learning from the best. And so I wasn’t really progressing. And also I wanted to start my own business and I wasn’t doing that. I was helping other people with their companies. So by getting caught up in the rat race, I basically wasted six years of my life.
Why did you stay too long?
I was constantly told, ‘Oh, you’re the youngest X to do Y in this company’ and so I got caught up in that. After school, you get trained to think that kind of stuff matters. But it really doesn’t. So I felt like I was progressing because according to some totally arbitrary metrics, that’s what I seemed to be doing. But I wasn’t building amazing new stuff that no one’s ever seen before. I was just really padding along in a very safe kind of environment.
Read part two of Jeremy’s fascinating story here.