Olusegun Ekoh

An AI-powered multi-agent system that automates job discovery, analysis, and personalized application material generation using CrewAI, LangChain, and Google Gemini.

• •Multi-agent architecture with specialized agents for job search, analysis, content writing, and outreach
• •Automated generation of customized resumes, cover letters, and LinkedIn messages
• •Integrated third-party API for real-time job discovery and application tracking
• •Robust resource monitoring system to track token consumption and operational costs across multiple model providers

A Retrieval-Augmented Generation (RAG) system that enables intelligent document Q&A. Upload PDFs or text files, extract content with OCR support, and get context-aware answers powered by OpenAI embeddings and GPT.

• •1536-dimensional embeddings using OpenAI's text-embedding-3-small model
• •Semantic vector search powered by PostgreSQL + pgvector extension
• •Advanced OCR support for scanned PDFs using PyMuPDF and Tesseract
• •Background task processing for document chunking and embedding
• •Context-aware responses via GPT-4o augmented with retrieved document chunks

An interactive PDF reader leveraging LangChain and OpenAI GPT that allows users to upload documents and engage in AI-assisted conversations to extract insights and answer questions.

• •AI-powered question-answering system with conversational memory
• •Semantic search using vector embeddings and Chroma database
• •Source citation with embedded PDF viewer for context verification

A semantic search engine that retrieves research papers based on meaning rather than keywords, processing 41K ML papers using transformer embeddings and GPU-accelerated similarity search.

• •High-quality 768-dimensional embeddings using MPNet model
• •GPU-accelerated indexing via Faiss for near real-time search
• •Embedding generation in ~41 seconds on GPU (~30 iterations/second)