John Oladokun

Aspiring Software Engineer and recent Computer Science graduate with hands-on experience building AI-powered security systems, automation tools, and cloud-based environments. Proficient in Python, Java, SQL, and virtualization — with strong grounding in data structures, algorithms, and system monitoring. Eager to join JPMorgan Chase’s Emerging Talent program to contribute to high-quality, scalable software solutions while expanding technical and collaborative skills.

• AI-Powered Dynamic Security Monitoring System, Towson University, 05/24, Developed an AI-driven monitoring system designed to detect abnormal behaviors in system processes for enhanced cybersecurity. Utilized machine learning models, including Isolation Forest and Autoencoders, to identify anomalies based on real-time data collected from system logs and event monitoring tools. Integrated real-time data streaming technologies and created alerts for abnormal activities to ensure proactive security measures. Tested the system with both simulated normal and malicious conditions, achieving high detection accuracy while minimizing false positives. Presented findings and outcomes, contributing to Towson University's research on AI and security monitoring.
• Cosmos Validator Node Assurance, Towson University, 09/24, Developed a dynamic web application to simplify node configuration and monitoring for non-technical blockchain validators. Utilized virtualization platforms like Proxmox to create and manage validator nodes, ensuring proper configuration, security, and uptime. Integrated node metrics using AWS tools to track performance and compliance, enhancing the reliability and security of the blockchain network. Automated security audits and smart contract registration using Cosmas, ensuring validators meet essential requirements.
• AI Powered File Organizer, 05/25, Developed a Python script to automatically sort files into folders based on file type. Utilized os and shutil for file handling and directory management. Improved workflow efficiency by automating manual organization tasks. Built an intelligent file organization tool using NLP embeddings to classify and auto-sort files by content. Implemented text extraction, cosine similarity scoring, and confidence-based routing to group files into context-aware categories. Improved organization efficiency and searchability through automated semantic classification and dynamic folder structuring.