Objectives of the service
Helping organisations to hire top diverse talent that is personalised to the company’s culture for today and tomorrow (futureproof), without bias. MeVitae automates the candidate shortlisting process and learns recruiter’s patterns to foster intelligent decision making, tailored to company’s needs. The solution’s steps are:
- Matching Algorithm: Scoring, ranking and shortlisting candidate CVs to determine job suitability. This uses past employer data, present (current CV and job spec) and future (via labour market trends). Understanding the labour market enables employers to understand talent marketdemands (e.g. what jobs are in demand within a certain demographic, current wages etc.) for future work trends
- Redaction: Using natural language programming (NLP), MeVitae removes all information from a candidate’s CV that could be cause for discrimination, e.g. age, gender, etc. before forwarding scored CV’s to employers
- Learning: Track employee progress for learning hiring behaviours. Over time, MeVitae becomes smarter. It learns more about the employer as they select candidates from the shortlist, therefore performing better the more it is usedand therefore ensuring that talent fits the company culture
Behind MeVitae’s technology is a breadth of data including the ability to be sensitive to talent market demands.
Users and their needs
Buying power for the solution is located with the Head of Diversity and Inclusion, HR director or at the executive team. However, the solution will be used by recruiters within organisations (UK and U.S. focus). Such organisations are:
- Medium and large companies with >250 employees
- Recruitment agencies as they often work with medium and large companies, and received high applicant numbers
- 3rd party specialist recruitment service providers including job boards and ATS providers
Their user needs are a solution that:
- Fit seamlessly into the current HR function (i.e. Application Tracking System or ATS)
- Needs to be personal to every company and their HR department
- Not have any biases (unconscious or algorithmic)
Faster than current processes in reviewing CVs; currently one CV every six seconds
Service/ system concept
MeVitae is seamless and all results are outputted within a client’s Application Tracking System (ATS). The integration with the ATS is seamless. The two types of information delivered are:
- Redacted CV that removed biases based on different parameters requires are interested in (e.g. age, gender etc.)
- Accurate and fast shortlisting of top candidates within ATS
Space Added Value
Using the ESA Patent 572 ‘method and apparatus for monitoring an operational state of system on the basis of telemetry data’, MeVitae will be able to analyse immense amount of unique time-series recruiter patterns data, specific to each recruiter, per job, per firm. This will help develop a unique selling proposition for companies by generating insights, picking up unconscious biases (causes behind lack of diversity in workforce) and predicting real-time trends in the labour market sector. The benefit of the ESA Patent 572 specifically is that it can be used for many parameters, including those MeVitae could potentially miss with little pre- knowledge, and is more efficient than threshold-based techniques. The Patent will be directly incorporated into MeVitae’s solution.
Current Status: https://www.youtube.com/watch?v=UhIOMGHrdlE&feature=youtu.be
- Developed Python code to extract organization names, job titles and qualifications from CVs. This is the result of developing and testing many potential solutions before settling on one that worked.
- Built databases of organizations, job titles and qualifications from around the world, which are needed in order to extract the data.
- Began testing ideas on how to extract information from job descriptions.
- Moved testing code for extracting University Name and Course from CVs from Python to C# so it can be used in production on the MeVitae Dashboard
- Made the MeVitae dashboard more modular to allow individual teams to work on specific areas without needing knowledge of the whole system.
- Reached out early adopters that will be taking part in the project
- Devised user needs survey to collect requirements based on early adopters’ feedback