Resources

Lessons in Deep Tech recruiting

This quarter I joined the Portfolio Talent team at Octopus Ventures after spending the past three years as a Recruitment Manager at DeepMind, one of the world’s foremost AI companies. Variety, as they say, is the spice of life, and it’s what attracted me to Octopus Ventures. I’m here to work with a wide range of super smart founders and co-founders; to help them grow their teams, and, critically, to bring my experience in the deep tech space to bear on their own journeys, helping them avoid some of the pain points and pitfalls I’ve come to recognise as common. 

I began my career in recruitment, where I found myself working on searches for analysts and statisticians, but over the span of a couple of years the landscape shifted. People started calling themselves data scientists, and big data and machine learning began to emerge out of a research space and into the commercial sector. I saw this change as an inflection point – and began to specialise.

My interests in AI research and development, and machine learning, saw me take a role at QuantumBlack – just as the company was acquired by McKinsey. It had been a boutique outfit – I was only the thirty-seventh employee, but its acquisition sent growth plans into overdrive. The briefs were fascinating, as clients began to realise the potential impact of machine learning. From optimising F1 teams pit-stop times, to using data to identify pain points in chip manufacturing, the teams we built had to adapt to different subject matter whilst applying statistical models to better predict outcomes.

With a newfound mandate for growth, we hired 100 individuals in the first year, from territories around the world. My role offered me carte-blanche, allowing me to develop interview processes, best practices for hiring and organisational design. I began to recognise the idiosyncrasies of recruiting talent in the space, and realised that developing my own in-depth understanding of the highly technical work I was hiring for, was crucial. The highly international recruitment process also presented a learning opportunity. I discovered more about the diverse talent markets and the different expectations of candidates, and developed a sympathy for the pace at which they operated. In Boston, US, for example, the window of opportunity to recruit talent graduating from MIT sat at around the two-week mark. Some candidates held offers from three or four other recruiters – from FANG-sized companies, to startups. We refined our recruitment process to better reflect the day-to-day work prospective employees should expect to be doing, introducing them to members of our data science team and made a feature of our USP: the sheer variety of the work on offer.

To many in the sector, DeepMind had acquired something of a ‘Holy Grail’ status, as an AI Mecca. It took me around three months to secure my role – and some 9 interviews; changing that unsustainable interview process was high on my to-do list when I started. My thesis, that to attract talent you have to understand the incredible (if often incredibly complicated) work they do, was reinforced at DeepMind.

I moved from research into the applied side. The company was working closely with Google, taking academic and research work and applying it to real-world settings and problems. The tangible impact this revolutionary technology can have, for communities and businesses, quickly became my real centre of interest. It also started to shape my hiring practices, as we became more focussed on attracting talent with an interest in the application of their research. 

As the implications of this technology became clearer to the commercial sector, people started looking around for ways to bring academia into the fold. Some companies made a point of hiring straight out of academic institutions, recruiting applied statistics PhD candidates. Others opted to retrain their own data analysts, to better equip them with the tools they needed for the kind of in-depth modelling work that this emerging field required. There was some debate on which is better, but there’s no secret recipe.

The speed of change in the sector has been incredible to watch first-hand, and given me an invaluable insight into some of the key pain-points that individuals and companies come up against.

The foremost of these is the tension between tech experts and commercial leaders. While the former may be motivated by a pure love of research, the latter have a perfectly understandable preoccupation with commercial outcomes. It’s a near universal friction point, but one I feel equipped to deal with.

1. Culture is critical. To navigate the potential friction between researcher and commercial teams, it’s critical to incubate a culture that gives space to researchers. At the same time, there needs to be a structure in place to help guide that research in the right direction, and they need to be encouraged to understand the commercial side of their field. 

2. Diversity is everything. More than just a force for social good, a broad range of ideas and experience need to be brought to bear on developing the technology. This is critical to insure against inbuilt bias, which will become increasingly important as AI systems become more widely adopted. 

3. Learn! Talent in this space is highly competitive. It’s not enough for a recruiter to try and charm an attractive candidate with buzzwords and a competitive pay packet; good recruitment is an ability to tap into their interests, motivations and passions. It’s therefore important to develop a deeper understanding of what’s going on in the space. 

4. And encourage learning. No one in the space wants to feel like they’re stagnating and with technology developing at an exponential rate, it’s important that talents are supported in their development, with a company culture in which learning is the norm. 

5. Be adaptable. When interviewing for these roles, you’re testing for highly specialist skills. There’s a broad sweep of neurodiversity across individuals in the deep tech, and companies themselves are in a state of positive flux, adjusting constantly to accommodate new developments. It’s important that all these considerations are taken into account when iterating on established processes – or creating new ones. 

The Talent team at Octopus isn’t just about offering operational input. I see my role as a chance to build and develop relationships, to offer support and to provide actionable insight and plans for growth, talent-brand building, or whatever might be needed. We don’t just invest in businesses, or a product; we invest in people – and it’s the people that make a company. 

Ultimately, you have the right mix of talent, I believe that great things are inevitable. My role is to help our portfolio find that perfect blend of talent, building cohesion and a healthy company culture, and creating a roadmap for the future.

Our portfolio company jobs board is live and we’d also love to hear any comments and thoughts on talent, culture and values. You can reach me at [email protected].

See more guides