Interview : Recruitment professional matches tech graduates to jobs using algorithms

Joe Afif, founder & managing director of graduate recruitment jobsite Gradit
Joe Afif, founder & managing director of graduate recruitment jobsite Gradit

Joe Afif, founder & managing director of graduate recruitment jobsite Gradit, is a self confessed “millennial“ whose big idea was sparked while working in recruitment sector.

He realised that there was a lot that could be improved in the recruitment process with data manipulation and the use of social media platforms.

Graduating straight into the recruitment sector in 2010, the 26 year old saw the sector change as LinkedIn was becoming a more common tool in the process.

He says social media was still seen as a fad in recruitment and digital recruitment systems were thought of as process bottlenecks and an annoyance to a generation of recruiters who were used to flicking through CVs.

Joe’s idea for Gradit was sparked while working at Dyson which he says was a fantastic place to work. He said “It is a company that truly hires the best and allows its staff the freedom to innovate, iterate and disrupt preconceived notions of how to do your job.”

He says this culture definitely helped to refine his business idea.

Joe saw an opportunity and launched Gradit, a software tool that makes it easier to hire the right entry-level talent in the tech sector.

Gradit uses unique technology which matches roles to students, so that they can apply for the right jobs quickly, easily, and effortlessly. This in turn means employers attract the right students and therefore make the right hires.

The Gradit homepage
The Gradit homepage

Give us an example of what a recruiter and a graduate need to do on the website in order to drive the process.

Often, websites in this space are simply advertising boards for employers to post jobs, and it's up to students to check in on the site every so often, sift through their jobs, and click on the ones that look right for them.

This is what recruiters call a ‘spray and pray' approach. It doesn't really work anymore, mainly because it’s labour intensive and boring. Students live in a world where convenience is king and we need a recruitment technology solution to match that.

Gradit works a little differently. Students sign up to the site and create a profile allowing them to input the exact data employers want to see when screening a candidate (education, technical skills, code etc).

Employers post jobs on the site. The system's algorithms figure out which profiles are the best match for their jobs and lets the students know about them. This means students only have to view and apply for jobs that are relevant to their skillset, which in turn ensures employers only attract the best-suited students.

It works the other way too. Employers can filter the database to only show the ‘Best Match' students for their jobs, and can contact them directly. We're seeing employers follow this route more and more.

In short, the tool cuts a lot of the noise out of the recruitment process to make it easier for students and employers to find one another.

So how did you go about developing the algorithm for this and creating the software that makes the system work?

It's quite simple really. Part of a recruiter's job is to (quickly!) look at an applicant’s CV, assess it against the job description of the role he/she has applied for, then work out whether the candidate is fit for the role and to what degree.

Our system does this through its algorithms. E.g. If a job description looks for X Y and Z, and the candidate has X and Y then the candidate will receive 2/3 and therefore a 67% fit for the job. Of course, this is oversimplified. There are many more variables and they're scored a lot more intelligently, but you get the picture.

I initially developed this over several months on an excel spreadsheet. It was a process of trial and error, immense iteration, but I finally got the algorithm that worked and matched with real life scenarios. In fact, they sometimes surpassed real life examples in situations where people overlooked the data in CV's and/ or job descriptions. With people, there is also the major issue of conscious and unconscious bias impacting a recruiter’s decision making. With the algorithms, the process is completely objective. Nothing is overlooked and it squeezes out bias.

People use the word “algorithm” a lot these days so it is interesting to see how one goes about creating one. How did you take this to the next stage and build it into the website?

So an algorithm is just a set of rules a computer needs to perform a task. I created the logic (the X, Y and Z example I gave) which a human could understand, however being a non-technical Founder I needed someone to translate this into what a computer could understand. To do this we hired software developers. Thankfully I spent a lot of time clearly writing out the logic so it was surprisingly quick for the developers to translate the logic into code, and thus develop our algorithms.

How are you going about marketing and attracting interest in your service?

We have two ends to market to – students looking to work in the tech sector, and employers looking to hire them. With regard to students, we work with a number of universities to promote our brand in the right circles. We have also done well through digital marketing.

On the employer's side, again digital marketing works well however we also like to be seen at relevant trade shows and events. I suppose the dream for every business is to generate leads through word of mouth and referrals because the product is so good – and we've seem some of that already which is fantastic!

How is it being received by your target market?

It's working well because it's very niche. Students who register tend to use the service until they've found a job through it. Usually traditional job sites and the like are just one of many and the user barely remembers what he/ she has signed up for, so the fact that we're retaining student users is fantastic. Employer feedback has been really great. We've had some of the biggest blue-chips in the country using the tool, and all of them are really supportive about it and gained real value through using it.

Are you self funded so far? For future growth will you be seeking outside funding?

Not anymore, we secured our second round of investment in August 2016. We're gearing for our 3rd round in Q3 as we're developing a significant new feature which will will completely shake up how employers engage and hire students and graduates. We're unbelievably excited about executing it… so watch this space!

Is Edinburgh a good place to base a tech startup? Does it have the right infrastructure? Which organisations have you found useful?

Edinburgh is a great place for a start up. It has a bit of that London buzz but on a much more manageable scale. It's big enough to attract bright people, creative technologies, and money; but small enough to have a tight community, and its an enjoyable place to live. I personally find Scottish Enterprise to be extremely useful, they've helped us a lot. StartEdin & ScotlandIS are doing some good things too.

What challenges have you faced?

The biggest challenge we faced is what every tech company faces – finding the right developers.

What one piece of advice would you give to someone thinking about a tech startup?

The time you have pre-startup is invaluable. You have the luxury of having a lot of time that you won't have the second you kick off. Use this time to do lots of primary research. Speak to lots of potential customers, find the right suppliers (and staff if you're at that stage), go to events and research the industry as much as possible. It'll stand you in good stead down the line.