When it comes to self-driving cars, AI usually grabs the headlines. You'd never have believed it 10 years ago, but advanced computer science is now sexy. 

I like to think of AI as the super-star driver of an F1 team. Having Lewis Hamilton or Fernando Alonso on your side increases your chances of winning the race, but without a top-class support team your expensive driver is wasted. In self-driving cars, the support team takes the form of top-notch mechanical, sensor, mapping and routing technology. Without this, no matter how good your AI is, your self-driving car will be rubbish.

Looking at the big players in self-driving cars, it's not clear to me who's in the lead. Google may have the best AI, but probably doesn't have the data on routing that Uber has built up. Then there's a host of other big players, from Apple to big car manufacturers, with more obvious gaps in their self-driving teams. This points to opportunities for start-ups to bring specialist skill sets to the table and quickly become a key player in an important strategic project for a big company. From better sensors to super-detailed mapping, there are some really interesting opportunities out there that would have seemed too niche or boring to invest in a few years ago.