EY India · Build
A candidate matching and attrition-risk engine, later productised by the client.
01
The situation
A large professional services organisation was matching candidates to roles through keyword search over résumés, which surfaced the wrong people and missed the right ones. Separately, they had no forward-looking read on which staff were most likely to leave.
Parameters in the matching engine
Models in production
Governed talent model
02
What we built
We built a matching engine scoring candidates against roles on more than thirty parameters — skills, adjacency, project history, tenure, seniority shape — alongside an attrition risk model trained on the organisation's own workforce history. Both ran against a single governed model of person, role, skill and engagement rather than against the résumé text.
Client took it forward.
03
Where it stands
EY subsequently productised the work internally as TalentMiner. That commercialisation was the client's own doing, not ours; what we built was the matching and scoring engine underneath it.
Handover status — Client took it forwardEY's own team took the model forward and productised it internally. We do not operate it.
Java · Spring Boot · React · Python · PostgreSQL
Next