AI Training & Research Lead (Final Year Project)
TextFine Tuning
Designed and refined a Network Intrusion Detection System (NIDS) by processing and labeling large-scale network flow data for machine learning training. Utilized class imbalance resolution techniques like SMOTE and ensured data quality for optimal AI model performance. Led the AI training process using TensorFlow and systematically evaluated accuracy and detection thresholds.
• Processed and annotated network intrusion event data from CSE-CIC-IDS2018 dataset.
• Applied synthetic over-sampling to balance data classes for better AI learning.
• Performed iterative evaluation and tuning of AI outputs for network anomaly detection.
• Oversaw end-to-end data preparation, cleansing, and labeling necessary for deep learning models.
Designed and refined a Network Intrusion Detection System (NIDS) by processing and labeling large-scale network flow data for machine learning training. Utilized class imbalance resolution techniques like SMOTE and ensured data quality for optimal AI model performance. Led the AI training process using TensorFlow and systematically evaluated accuracy and detection thresholds. • Processed and annotated network intrusion event data from CSE-CIC-IDS2018 dataset. • Applied synthetic over-sampling to balance data classes for better AI learning. • Performed iterative evaluation and tuning of AI outputs for network anomaly detection. • Oversaw end-to-end data preparation, cleansing, and labeling necessary for deep learning models.
2026 - Present