Habeeb Ajibola

Dynamic data analyst with a proven track record at ICARE Clinic, leveraging advanced AI techniques and deep learning frameworks to drive actionable insights. Expert in data quality and cleaning strategies, I excel in synthesizing complex data into impactful visualizations, fostering collaboration, and enhancing decision-making processes. Passionate about innovation and continuous improvement.

• AWS Cloud Practitioner
• AWS Machine Learning Associate
• AWS Solution Architect Associate

• Ai-Powered Multi-Modal Deepfake Detection System, Developed and trained modality-specific detection engines—Vision Transformer for image analysis, ResNeXt-LSTM pipeline for video classification, and fine-tuned BERT for text—achieving F1 scores above 0.90 across modalities., Architected a cloud-native data pipeline on AWS, leveraging S3 for scalable storage, AWS Glue crawlers for automated schema cataloging, and Athena for serverless querying of over 500K images and 1,000+ videos., Developed user-centric visualization tools including a Graido interface for real-time testing and Power BI dashboards with Grad-CAM and temporal saliency maps for model interpretability.
• Gen Ai Symptom Checker Application, Designed and implemented a retrieval-augmented GenAI system using LangChain and Pinecone to ground LLM responses in a structured medical knowledge base., Built and fine-tuned urgency-classification and symptom-inference models on a synthetic SQLite database of 5K+ patient records and 15K+ symptom reports, incorporating PII scrubbing and prompt engineering., Developed a Streamlit chat-style UI enabling conversational triage, achieving 75–80% overall accuracy and 100% recall on critical emergency scenarios during evaluation.
• Ai-Driven Analysis of U.S National Parks, Collected and processed 20,000+ data points from the National Park Service API using R, performing data wrangling and feature engineering for predictive analysis., Developed and optimized machine learning (GBM) and deep learning (H2O) models to forecast visitor engagement and park popularity., Conducted feature selection, correlation analysis, and model tuning, delivering data-driven insights on visitor trends and seasonal demand to enhance park resource allocation and operations.