Grow Your Own Experts
Stardate 95544.2. Captain’s log supplemental. It’s been nine months and no good candidates are passing through. The two who were a match turned down our offers. Meanwhile, we stagnate and the competition is killing us. We need experts stat! I don’t know what we’re doing wrong… our job description clearly lists the exact skills, tools, knowledge, and experience needed. We are losing hope…
If you find yourself in this predicament, consider this: your hiring criteria may be off and your candidate pool may be too narrow.
When you focus on exact matches, the talent / candidate funnel is quite unforgiving: there are very few people out there with the perfect skill set overlap to begin with. Fewer of those also have the exact experience you (think you) require. Fewer still are good at what they do, smart, nice to work with, diligent, etc. And even fewer prefer your company over all of their other options.
If you don’t like these odds, perhaps it’s time to adjust your criteria and broaden the candidate pool. Find smart, dedicated, driven people with a strong foundation. Extend them the opportunity, teach them, let them teach you, and they’ll become the experts you need in no time.
So how do you broaden the candidate pool in practice? Glad you asked. Here are just three of the strategies I’ve successfully used in the past.
Skills: Tech Can Be Learned
Find people in your field with expertise in a slightly different area — they will initially have only a subset of the skills you require.
In the beginning of 2009, I was starting to build BlueKai’s engineering and analytics team, responsible for designing and developing the company’s big data platform. In just a few years we scaled to over 2 billion unique user profiles and to tens of trillions of events and transactions per month. It was the early days of commoditized, massive scale, distributed systems and at the time Hadoop was the way to go. Without thinking twice about it, I set out to hire experts in this ecosystem, namely “Hadoop engineers.” Silly me.
The problem was that except a for few folks at Google and Yahoo, there weren’t many “Hadoop engineers” out there. I searched everywhere, left no stone unturned, and in the process rejected many amazing back-end and full-stack engineers because of their lack of experience with Hadoop.
After nine months of failing miserably, I realized that my mistake was equating (a) success in the role of platform engineer with (b) knowledge of a specific technology. I had been looking for people with Hadoop experience, when I should have been looking for strong engineers with the passion to learn and the focus to deliver.
This experience radically changed the way I approached hiring: recognizing that skills can be acquired, considering a broader candidate pool, and identifying true predictors of success in a given role. That year I hired some of the best people I’ve ever worked with — none of them Hadoop engineers. They became experts in no time and together we built one of the best platforms around.
Foundation: Find Knowledgeable Neighbors
Broaden the pool even further to include experts in neighboring fields with a solid foundation, who have proven they can acquire skills such as the ones you need.
During my Ph.D. I went through a couple of transformative internships at IBM TJ Watson, in the applied machine learning (ML) and natural language processing (NLP) group. This is the group that earlier had laid the practical foundations for modern statistical machine translation in the late 1980s, revolutionizing the field. The same group produced the large scale NLP, information extraction, and overall machine learning behind the Jeopardy-winning Watson. As I started to interact with scientists and engineers, I noticed that many of the NLP, speech generation, and ML experts had formal backgrounds in electrical engineering and physics, rather than computer science. Not only were they great at their craft, but they also had perspective, coming from a different field.
Years later, I started my own company, Level Up Analytics, together with a couple of good friends, both of whom also have Ph.D.s in physics. We ended up hiring quite a few people with backgrounds in electrical engineering, physics, astronomy, space science, economics, etc. These were some of the best engineers, data scientists, and product managers I’ve seen, all successful in their roles. They built their computer science careers on a foundation of math, engineering, critical thinking, heavy-duty data processing, and coding. On top of that, they also had the passion and the drive to build and deliver good products. As for tools and technologies, they quickly and thoroughly learned what they needed along the way.
If I learned anything from these experiences is that it pays to broaden the pool of candidates by considering neighboring disciplines. For example for data science consider candidates who are hands-on Ph.D.s and postdocs in math, sciences, and engineering fields. They have the math and statistics background, the focus, and the depth. If you also test for practical problem solvers with good coding skills, you’re already ahead. For data engineering, consider looking at electrical engineers and embedded systems engineers — they already love optimizing, performance, and depth.
This really works. Insight Data Science, a wonderful company I advise, proves this every day. Its fellowship program identifies and brings powerful and creative minds from neighboring disciplines into data science, data engineering, health data, and AI. They go on to become key experts at startups, as well as established companies. In a sense, for data-related tracks, Insight helps you tap into a broader the pool to find great candidates, while reducing your effort and risk.
Geography: Zoom Out
Broaden the pool to include experts who live somewhere else— either bring them to you or take the job to them.
Before the ink was dry on my diploma, I was recruited by a quasi-startup, a visionary, move-mountains kind of group, within Siemens Healthcare. We were asked to bring intelligence — i.e. machine learning — to products and see them through to market. A couple of projects were were particularly ambitious: one on faceted search for medical records and one on active learning for information extraction from noisy documents. With limited resources and a strong need for data scientists and engineers, we iteratively broadened the candidate pool. Soon after, the team expanded to: front-end engineers in Princeton who learned new, advanced libraries on the job; app developers in Bangalore who had never before worked in search; medical domain experts in Philly who did not have prior experience in relevance ranking; back-end engineers in Mountain View, who relocated from Germany and took on the novel task of developing deployment services for machine learning. Fast-forward to today and many of us work together, now at Yoyo Labs.
Taught (not very gently) by scarcity and necessity, I ended up using geography to my advantage in every team, finding or growing experts across the US (Bay Area, Seattle, Philly, Boston, etc.) and internationally (India, Germany, Australia, Venezuela, UK, New Zealand, etc.).
And why not look across the world to broaden your candidate pool? Either create the opportunity for incredible people to relocate and join your merry band, or learn how to build, nurture and manage distributed teams. We’re currently doing both at Yoyo and looking at the people we work with, we couldn’t be happier with our approach.
When it’s all said and done, the opportunity cost is too great to be comfortably passive. Find smart, creative, and collaborative people with a solid foundation and then inspire them and watch them soar. Grow your own experts!
___
If you enjoyed this post, click the👏 below so other people can read it here on Medium.
Yoyo Labs is a premier Data & AI consulting firm. We specialize in building custom, state-of-the-art, high-impact data solutions. Our clients span verticals (Advertising, FinTech, Healthcare, Social, Manufacturing, Mental Health, etc.) and sizes (nonprofits and startups to Fortune 100). We care about product and we bring our deep data expertise and a relentless quality focus to bear. The more complex, large scale, intractable, gnarly data product problem you have, the better! It’s right up our alley. Talk to us! We can help.
Lucian Lita is founder of Yoyo Labs, previously founder of Level Up Analytics and data leader at BlueKai, Intuit, and Siemens Healthcare.
