Project Description

HR software for the public sector, is currently seeking a remote contract
Data Scientist located in the US or Canada time zones. The Data Scientist will work closely with
a lead Data Scientist on the Business Intelligence team and play a pivotal role in augmenting
product functionality by processing and categorizing user inputted text.

This is a text processing problem. We need to analyze jobs and candidate resumes, categorize them, find common terms and determine the level of compatibility between applicant and a job.

Responsibilities

• Preprocess and clean noisy user-inputted, unstructured text
• Apply pre-trained language models or train new models for text embedding
• Assess quality of labeled training set to establish ground truth
• Train classification models for predicting predefined categories
• Create new subcategories where needed using clustering or similarity matching methods
• Assess quality of model classification and make iterative improvements throughout the
  entire modeling pipeline

 

Requirements

• 3+ years of Python experience, including pandas, NumPy, and scikit-learn

• 2+ years of Natural Language Processing (NLP), specifically various text preprocessing,
  embedding, and feature engineering techniques

• Knowledge of pre-trained language models

• Experience with classification model (SVM, Random Forest) and/or clustering models
  (hierarchical clustering, k-means, KNN)

• Experience data mining public data to enrich training set through APIs / web scrapers

• Experience collaborating on analytics, data science, machine learning, or NLP projects

• Experience working with large datasets

• Experience contributing to predictive models deployed into production