Developed an intelligent candidate ranking algorithm that enables recruiters to assign weights to desired features for merit-based suitability scoring, promoting unbiased candidate evaluation. Leveraged a RAG based large language model to extract key candidate features from resumes and jobs, after conducting surveys with recruiters, informing algorithm development for automated role-fit assessment. Integrated algorithm outputs with organizational DEI initiatives, presenting diverse and qualified candidate pools aligned with equitable representation goals.
Retrieval augmented generation (RAG) . LLM . LangChain . NLP . Python . Fine-tuning
01.
Understanding sentiments within tech communities on YouTube, particularly focusing on discussions surrounding Apple and its competitors. By collecting video transcripts and comments, the project offers insights into unfiltered consumer opinions and sentiments
Social Media Analytics . Web Scraping . LDA . Aspect-Based Sentiment Analysis . Python
02.
Utilized ELSA's dataset to analyze workplace sexual harassment trends, uncovering high tolerance levels and low reporting instances. Explored training's impact on awareness and identified significant implications on employee well-being and productivity.
Python . Tableau . Hypothesis Testing . NLP
03.
Predictive model to estimate AirBnB prices in popular European tourist destinations by analyzing correlations between various factors such as location, property type, and amenities.
Python . Random Forest Regressor . XGB Regressor
04.
Data-driven initiative to optimize LinkedIn Premium features, with a primary focus on attracting and retaining students and early-career job seekers, who represent the platform's second-largest user group.
Python . Tableau . Thematic analysis . Statistical analysis . Hypothesis Testing