Advice to a fresh graduate for getting a job in AI/ Data Science

Author: ajit jaokar


After a recent webinar, I was asked about advice for getting a job in AI for a fresh graduate


This is a good question and not often answered

Here are my thoughts



  • Firstly, AI is a vast topic. Everyone has a limited view on AI based on their personal experience. And that includes me .. So, you should seek multiple viewpoints on this question.
  • Also, everyone’s perspective comes with a personal bias. I am biased towards engineers. In my view, if you do not come from an Engineering / Masters / PhD background, AI (at senior levels) will be hard.
  • Of-course you should have a strong coding background, have your own github repository, be familiar with maths etc. I will not repeat these points here
  • There is a debate about online education in AI. My view is: if you have taken steps to learn AI – online or offline – you are ahead of most other people. So, I am not biased to online education in AI
  • I see AI jobs to be actually three job categories: Data Engineers, Data Scientists and Devops. For this reason, you need an Engineering/Computer Science/PhD background (ideally)

With this background, below are my views and I hope you will find them useful


Advice to a fresh graduate for getting a job in AI/ Data Science

  • Make big bets: In our Oxford University Course on Data Science for IoT, Jakub Langr covers GANs. Jakub has one of the best books on GANs I have seen. When I had lunch with Jakub in London – he mentioned something interesting. Three years ago .. (that’s an age in AI!) he saw the potential of GANs and decided to bet his career on it by reading every paper, article etc on GANs. Three years later – he has a book on GANs and due to it he teaches at Oxford on GANs. This strategy is actually rather unusual. Not many people make big bets .. they try too many small things and they fail to distinguish themselves in an area at a world-class level. I would prefer to see more young people to be technically ambitious. At the initial stages, of your career you really have little to lose!
  • Understand the market: Most people do not think of AI as a ‘market’ but it is. Like any market, it has its nuances and regional characteristics. For example, here in the UK AI jobs are booming – with very clear centres of excellence in London, Oxford and Cambridge. In Israel, we have Ben Guiron University of the Negev for Cybersecurity. In India Bangalore and also many such hubs in China. The point is, the you have a better chance in a place where a vibrant ecosystem already exists
  • Business knowledge: Business knowledge(sector knowledge/domain knowledge) is often under-rated. But as AI becomes more widely deployed, business knowledge matters. Invest in one industry(for AI) if you can
  • Big problems: Work with big problems. The bar for AI is raised. Its no longer enough to work only with sample datasets like Mnist if you want to distinguish yourselves. A good example of solving complex problems is Piotr Skalski’s blog 
  • Academia matters in AI: I have long been associated with academia and industry both. In AI, good academic collaborations matter. If you can, cultivate such relationships
  • Small company: If you can work with a small company/start-up who is doing real AI work – do so. You can learn a lot
  • Large companies: Go for companies which value AI like Microsoft, Google , Amazon, facebook, Nvidia etc. Not all large companies value AI. Many industries are still lagging
  • Hone you Research skills: You will spend a lot of time working with research papers.


Having said all this, AI looks glamorous from the outside


Once you start working, the other side of the AI jobs is often boring(ex: data cleaning) !


All the best ..



Image source:

Go to Source