Google and other tech giants often have huge recruiting efforts that can target students — but startups might not have those kinds of resources to find or grab them before they head off to those huge companies.
The biggest tech companies in the world can offer a lot of resources and bigger compensation packages, but startups may still offer a huge opportunity too. That’s why Edouard Harris and Jeremie Harris started SharpestMinds, a tool aimed at finding those students before they get snatched up with huge offers from the FAANG (Facebook, Apple, Amazon, Netflix and Google) tech giants of the world. SharpestMind aims to create a new kinds of machine learning-driven service that can identify tech talent early, get them in the door at startups, and eventually lead them to a potential hire for a problem they’re interested in — and not just the appeal of the massive scale that Google and others can offer. The company is launching out of Y Combinator’s 2018 Winter class.
“In our case, it came down to the fact that we started off as grad students, we saw the cream of the crop grad students getting picked off by Google or Facebook,” Jeremie Harris said. “They were getting them even before they graduate from a graduate or undergraduate program. It was such a struggle [for startups] because all the best people were getting poached early. Big companies that have giant campus recruitment operations. If you’re a startup, you don’t even get a shot at these people. They don’t even exist on your radar. What we figured was, why don’t we take what they’re doing for themselves, build it into a service, and expose that service to startups and companies.”
Startups (or smaller companies that don’t have the resources of a FAANG recruiting operation) post the job openings they have and the requirements they need, first. Then, as students start applying through SharpestMinds, the startup matches up their skillset with the requirements of the company that posted their job. Once that’s matched up through a number of tracking tools, they’re interviewed, and then placed to work on some problems early on. Think of it as a kind of trial run (or a set of practice problems) with a company that’s trying to get something into production and out the door. It’s not fitted directly to some sort of key-performance indicator, but it’s a good way to let startups get a feel for the candidate.
If that all works out, the startup gets early access to someone who they’ve already worked with and already knows the system. That could offer a huge edge in recruiting as not only does the startup determine if the candidate is a technical and cultural fit, but the student has an opportunity to do a trial run — and if they like it, they might have a preference over a company like a Facebook or Google. That startup might not offer the capacity to run a test against tens of millions of people, but they might be able to offer a value proposition for a problem that the candidate is passionate about, and then get them on board. The placement program still doesn’t encourage students to drop out, as is the sort of tradition in Silicon Valley.
“In practice, it depends on what the masters or Ph. D. is in,” Edouard Harris said. “It turns out that speaking candidly about my own background, Physics is not the best way to get ready for industry. It’s possible that if you’re doing a Ph. D. in physics, the transition to industry might actually be your best move. For some people we don’t people encourage directly to drop out. It has happened for one or two companies.”
That so-called switcher problem is one of the hardest to solve. The industry is often seen a starved for talent in a number of areas like data science and machine learning, especially as more and more companies look to devote resources to those parts of their technology stack. The above-mentioned Ph. D. in quantum mechanics may not have the technical background that a startup critically needs, but at the same time they may also have a number of pet projects and have the capability to learn the skills on the job — and do so quickly, meaning the startups can have a short training ramp-up for them. That not only gets them AI talent, but gets it before they are even on the radar of a Google or Facebook.
And that’s also part of the potential appeal for SharpestMinds among students and one that hopefully gets applications in the door, Jeremie Harris said. Getting real-world experience is often the best way to get started, especially if academia hasn’t caught up with the problems that startups now face that only apply a part of what a degree might entail. For Jeremie Harris, that a-ha moment came when a student passed by his graduate office, and talked about how they wanted to work with them on an image recognition startup because they just wanted to get their hands dirty. And when a student applies to a job and they don’t get it, SharpestMinds will reach out to the candidate that did get the job and they’ll do a sort of Q&A session.
To be sure, there’s a lot of competition in the recruiting space in Silicon Valley. There are tools like Headstart, which aim to analyze candidates and try to determine the best fit. As these companies get a bigger and bigger pool of data, they might be able to identify talent and that switcher group that has the skills but not the resume. But as both of them have backgrounds as graduate students, and a lot of first-hand experience dealing with the problem, Jeremie and Edouard Harris hope they can build up a big enough cohort of students that the solution will start to be more and more efficient.
“You have people that don’t know what it takes to get industry ready, or what should be next,” Jeremie Harris said. “It’s all over my [LinkedIn] feed, there are data scientists prognosticating how I became a data scientist, and people gobble this up. The amount of traffic dedicated to talk about what you should do to become a data scientist in insane. We realize that’s a reflection of the uncertainty of potential hires.”
