ASOR AHURA

AI/ML Engineer | Health Informatics Professional


About Me

Hi! I'm Asor Ahura, an AI/ML Engineer and Health Informatics Professional with a passion for delivering real-world AI solutions that drive innovation and efficiency.I specialize in designing and implementing intelligent systems that optimize workflows, enhance decision-making, and automate processesMy expertise spans machine learning, predictive analytics, and health information systems, enabling me to bridge the gap between cutting-edge technology and practical applications.I am committed to leveraging AI to solve complex challenges, improve operational efficiency, and create meaningful impact across industries.


Skills


Python | Tensorflow | Postgres | Oracle Cloud | Docker

  • Technical Support

  • Troubleshooting

  • Systems thinking

  • Analytical thinking

  • Team management

  • Communication


Projects


Cervical Cancer Screening Tool

The Cervical Cancer Screening Tool is a user-friendly, web-based app powered by YOLO, a state-of-the-art deep learning model, designed to analyze medical images for cervical cancer detection. Healthcare providers can upload cervical scans, receive instant AI-driven diagnoses with confidence scores, and escalate uncertain cases (below 90% confidence) to clinicians via email for further review. The app ensures seamless tracking with facility and client codes, prioritizes data privacy through temporary file management, and offers real-time feedback for easy use. Ideal for primary healthcare providers and specialists, it improves accuracy, speeds up diagnoses, and is scalable for resource-limited settings, making early detection and timely intervention accessible globally. Join us in revolutionizing cervical cancer screening.A demo of this app is available only on demand. Send a mail.


IMAGE TO JSON

Revolutionizing Data Extraction From Handwritten Forms

Image2JSON is an AI-powered document digitization solution that transforms handwritten forms into structured data using LLaMA Vision technology. The app addresses common challenges in manual data processing by offering automated text extraction, JSON structuring, and secure cloud storage through Supabase, while providing instant Excel report generation. With a 90% reduction in processing time and enhanced accuracy through AI, it serves diverse industries including healthcare, finance, education, and HR departments. Built on Streamlit with Python, the platform offers an intuitive interface that enables users to upload documents and receive structured data instantly, making it a cost-effective solution for organizations looking to streamline their document processing workflows.


RESUME REVIEW

Smart Matching for Smarter Hires

Resume Review is an AI-driven application designed to enhance the hiring process by automating resume screening and candidate matching. Utilizing Streamlit for a user-friendly interface and Groq's advanced AI technology, the app processes uploaded resumes and job descriptions to deliver insightful candidate analyses. This innovative tool aims to save HR professionals valuable time, reduce bias in hiring decisions, and improve the accuracy of candidate matches, ultimately leading to more effective and informed hiring outcomes.


DRIVER ALERTNESS PREDICTIVE MODEL

This project focuses on developing a machine learning model to detect whether a driver is alert or not alert while operating a vehicle. The model uses a combination of vehicular, environmental, and driver physiological data collected during driving sessions. Using XGBoost classification and extensive hyperparameter tuning, the model achieves over 93% accuracy in predicting driver alertness states.


PDF DOCUMENT Q & A SYSTEM

This sophisticated PDF Document Question & Answer System is a cutting-edge application that transforms how users interact with document collections. Built with Streamlit and powered by advanced AI technologies, it enables natural language interactions with PDF content through an intuitive chat interface. The system leverages Groq's Large Language Models and Google's AI embeddings in a Retrieval-Augmented Generation (RAG) architecture, ensuring accurate and context-aware responses. Documents are processed into vector embeddings for efficient retrieval, while the LLM generates natural, coherent answers based on the relevant context. The application features a modern, responsive web interface with real-time processing, comprehensive error handling, and a persistent chat history, making it ideal for research, documentation analysis, and knowledge management tasks.

© 2025. All rights reserved.