As Machine Learning (ML), a specialization of Data Science / Artificial Intelligence (AI), moves from a hype to becoming mainstream, there are still a lot of unknowns around the role ML plays in businesses. What exactly do companies hire ML for? What are the skills they need? As this article puts it 

Clarity around how data science and machine learning solve business problems is a lot less plentiful. If you ask me, that’s the real skills gap.

For instance, did you know the top hiring role in Data Science is no longer "Data Scientist". In fact, it’s relating to Machine Learning skills. 

As it is happening, at Gooroo.io, we are excited to be able to look at the live trends in front of us. In this post, we'll take you for some deep dives into the specific skills and roles insights in the ML and AI world. 

The emergence of Machine Learning Engineers

As above, do you know that the top hiring role in the data science profession is now "Machine Learning Engineer" (or its variants Machine Learning Developer, ML Software Engineer etc)? It is now commonly seen on job placements, especially in more matured tech locations (such as San Francisco Bay Area, Boston, London, Seattle). 

The tag cloud below shows the the prominence of such roles in the diverse ML roles landscape (size correspond to percentage). Jobs for Machine Learning Engineers roles now accounted for 10% of all ML jobs. 


Machine Learning Hiring Meccas

Our data reveals the top hiring locations for Machine Learning in 2017 as London, San Francisco Bay Area, New York, Cambridge/Boston (MA), New York, Seattle.

As expected, this list is still dominated by top technology and R&D focused locations. The majority of these locations are in the Bay Area - they have their own sub-clusters (San Jose, San Francisco, Santa Clara, Mountain View, Palo Alto,  Sunnyvale). Among cities, it is noted that London stood out because of the concentration of (tech) jobs in the UK (the second UK location is vastly further behind). 

Top locations hiring for Machine Learning (size corresponds to number of jobs) 

Hiring companies - from megacorps to start-ups 

We've seen hiring happening across the industry, from large organizations such as Intel, Facebook, Lenovo, to tech unicorns such as Uber, Udacity, Human Longevity (life science tech startup), Square, to emerging boutique startups such as X-Team, Wave, Splunk. 

It is interesting to see that companies hiring for ML now range from tech companies (such as Adobe, Facebook, NVIDIA) to professional firms (EY, JP Morgan Chase) through to brick-and-mortal retail (Home Depot), Guardians. 

Excluding recruitment agencies, top of the hiring chart are still tech giants, including Facebook, Samsung, Intel, Cisco. Large companies such as Google or Amazon

What are the sub-specialities that are hiring in ML?

Just like any technology that becomes pervasive, ML has formed sub-specialisations. We have identified the following sub-specialities which companies are now hiring for: 

  • Scala / Big Data
  • Go (ML) 
  • Azure / AWS
  • Computer Vision
  • Neural Network and Deep Learning
  • Sensors
  • Search and Machine Learning
  • Database (NoSQL)
  • C# / AI 

What are the functional areas hiring for ML? The productionization and commercialization of ML. 

Interestingly, jobs that mention "Machine Learning" are no longer confined to engineering roles. As the industry is becoming mainstream, it requires different functional roles to support, ranging from engineering to UX to sales. It is a signal that the industry is moving into "production phase", rather than just an R&D phase. The top functional areas with job openings related to "Machine Learning" are: 

  • Engineer
  • UX 
  • Sales
  • Marketing
  • Research and Development
  • Marketing
  • Project Management

Conclusion

Our data set provides some interesting insights about the state of Machine Learning industry (and AI in general). Firstly, it suggests that ML as an industry is maturing quite quickly, well beyond the R&D phase that many people still associate it with. While top opening roles are still within engineering domains, the industry is most likely entering a production/commercialization phase with hiring happening more commonly in other functional areas.

Secondly, the diversity of skills and specializations continues to surprise us - it is no longer "stats, maths, programming", but rather a range of skills and tools and techniques. It is an exciting time if you’re considering career development in this area, as demand (jobs) is ramping up so quickly. If you want to follow the latest trends and analytics on specific skills, sign up with us at Gooroo.io, or head to our "Skills & Salaries Analytics" page at https://gooroo.io/analytics.

Brad Nguyen

Lead Data Scientist, Gooroo in residence