Open AI Embeddings
Transformed raw skill texts into high-dimensional vectors using OpenAI's `text-embedding-ada-002` model. Applied cosine similarity to measure closeness between user skills and job roles.
My Signature Project
AI Skill Gap Analysis: Closing Internal Gaps for AI Readiness
My MSc project delivers a data-driven analysis of AI talent readiness, using advanced techniques like embeddings, cosine similarity, and clustering to uncover critical skill gaps across real-world AI job roles. The project features interactive Power BI dashboards that provide actionable insights for strategic workforce planning, talent development, and AI capability building within organizations.
Skill Cluster Mapping
This scatter plot illustrates the clustering of AI-related job roles based on their semantic skill similarity.
Each point represents a job or skill profile, encoded into high-dimensional vectors using OpenAI’s text-embedding-ada-002 model.
Dimensionality reduction via UMAP and clustering (e.g., K-Means) revealed hidden patterns and grouping structures within the AI job market.
AI Talent Skill Gap Dashboard
Developed an interactive Power BI dashboard that analyzes skill trends from over 1.3 million AI-related job postings across the US, UK, Canada, and Australia. The dashboard visualizes top emerging skills, co-occurrence heat maps, and semantic similarity scores across roles to reveal hidden talent gaps and in-demand AI skills.
AI Job Recommender (Live Tool)
Developed a skill-based AI Job Recommender using OpenAI embeddings and deployed it as an interactive web app via Hugging Face. Users can enter their skills and receive personalized AI-related job suggestions based on semantic similarity matching from a dataset of 50K AI job postings.
