Predikt.co aggregates and quantifies your professional data spread across the web, including your resume. Use the Predikt Score to evaluate how much of a fit you are for a particular job position. So far 200+ users have registered and scored their profiles.

The idea behind Predikt is to bring together all your relevant professional data and intelligently quantify it in such a way that the number would represent the potential fit for a specific job role. E.g. Predikt Score of 81 for the role of Data Scientist. The score is specific to the role and the most ideal candidate for this role would have a score of 100. For instance, top data guys like Peter Skomoroch will likely have a score between 92 to 100 for their respective roles. There is some more additional data which supports the score, we call it ‘Domains’. It can be defined as an area of expertise which is not as superficial as saying ‘Information Technology’ or as deep as something like ‘Hadoop’, but somewhere in between. Lets say for a person with data relate skills, the primary domain could be ‘Data Science’. There are also secondary and tertiary domains of expertise, in this case they can possibly be Software Development, Web Development etc..

The Prediction Algorithm:

So how is this Score calculated and what sort of algorithm is being used? The score is computed for each individual based on multiple factors such as skills, education, experience, projects, interests etc. But that doesnt sound very new, does it ? There have been some services which attempted to do similar stuff, so how is Predikt different? We predict the likelyhood of a person being a potentially good fit for a certain role. The way we are doing this is by using similar other profiles to deduce the factors which make a successful candidate and use them to predict potential candidates. e.g. People who have studied in university X, have Y yrs of experience and have Z skills go on and achieve position ‘A’. These factors can help predict someone who would likely land in a similar position.The approach is somewhat similar to how a Credit Score is computed or even a Moneyball approach.

As of now, linkedin and facebook integrations are functional, github is in progress, so stay tuned. While we are still figuring out what other networks to integrate and what factors to include, don’t hesitate to drop us a line on amogh@predikt.co or tweet (@click_amogh ) if you have any thoughts. There is a HackerNews thread which you can jump on with your views.

Dont forget to Get your Predikt Score and discover your areas of professional expertise. And dont worry, its completely free ! Let us know what you think. If you are interested in contributing or working on this product, just drop a line or comment.

(If you are not using adblocker, you’ll probably see an Ad below. feel free to ignore it and click the share button beneath it)

Advertisements