The client wants to save human efforts on the resume selection and evaluation process and use ML/AI for following cases to scale their business.
Classify resume on predetermined job roles.
Rank Resumes.
Best match based on Job description.
Integrate the new system with their existing resume hosting platforms like CRM, MS Dynamics etc.
SOLUTION
Our resume recommendation and ranking solution involves preprocessing of raw resume data, tokenization and normalization of text, and then passing it to different trained models for classification, resume matching, and ranking. For resume context matching and ranking we would be using LangChain framework along with pinecone Vector DB to store the embeddings. For classification, we are using supervised learning approach, Logistic Regression, and XGBoost. In the end, we are doing API integration with external apps to expose the resume classification and ranking to external apps.
It is broken down into the following high-level steps.
Document Processing
Data Parsing
Classification
Recommendation Engine
Document Ranking
API Integration.
BUSINESS IMPACT
1
Reduced Human efforts in the sorting and selection process and save cost.
2
Reduce interview rejection rate and improve hiring rate.
3
Improved Customer Satisfaction.
4
API integration extend potential target application and hence larger customer segment.