Computer Vision: The Transition from Academia to Industry


Computer Vision: The Transition from Academia to Industry

Posted on 07 December 2022

Computer Vision: The Transition from Academia to Industry

From facial recognition preventing criminals in prevailing to medical imaging advancing diagnostic and rehabilitation systems, Computer Vision has a drastic impact on the development of technology today. Those involved are collaborating to initiate the inevitable evolution of the world we currently know.

As a Machine Learning recruitment consultant at DS Group focusing solely on the Computer Vision field, I feel compelled to help with journeys up the ML ladder. I wanted to create an article that would provide vital information for those trying to break into the industry from academia. This article will include key points to help with cleaning up your portfolio, preparing for ‘the jump’ and giving you a brief insight into the mind of the employer.  

When approaching experts I felt little to no resistance on providing advice to help me with my article. Some clear themes began to emerge from the advice I had collated.  

The integration of industry into study was a big theme that continued to be evident in my discussions. Biswarup (Team Lead Manager) mentioned that an internship working on relevant projects would definitely boost a portfolio coming from academia and Paul (Head of AI) agreed. Guillaume (ML & Robotics Lead) included that evidence of the candidate continuously using and improving their software engineering skills is ideal too.  

Most experts agreed that research topics which are relevant to the employers projects and mission are a clear indicator to the company that you are able to do the work required. Relevant publications were mentioned by Biswarup, Guillaume, Paul and Severin (CEO – Founder), with Severin adding that the earlier on the publications are into the candidates academic career, the better. Biswarup highlighted the positive implications that personal projects have on your portfolio as well, especially when relevant. 

Another key theme that stood out was the ability to understand the practical implications and infrastructures of the real world. James Paterson (VP of Engineering) stressed that you must be able to understand, analyse and overcome failure. He also stressed that being able to deal with customer needs and the ability to explain a technical problem to a non-tech user is key to an industry role. This was supported by Violet (Head of Platform Ai) who mentioned that understand the infrastructures are essential when going into an industry role. 

Although academic excellence makes the shift much easier, it seems as though it is not absolutely necessary to find an industry role. Violet made a clear point that she didn’t take the traditional route and thinks that if you don’t have a relevant Msc or PhD, it’s not the end of your career as many employers do hire off of excellence. Severin added that he prioritises talent over qualifications during although studying at a top tech institution makes things easier, almost every expert agreed its not completely necessary to land your dream in the field. Ultimately it's evident that talent always shines through on a portfolio. 

To wrap this article up, to me it seems that the success of your integration from academia to industry is based upon both your ability to add to the project and understand the world of industry and the problems/change that comes with it.  

If you master your ability to understand, analyse, develop and overcome, you are very likely to ease your step into industry.  

I would like to thank everyone that helped provide the content for this article. If you have questions about getting into ML please do get in touch!

Share this article