🧠 Can Large Language Models revolutionize network security? Taeyang Kim thinks so—and he's got the proof. In this forward-looking BSidesSLC 2025 talk, Taeyang Kim (Machine Learning Engineer at Pattern Inc.) introduces an innovative full-stack LLM-based Network Intrusion Detection System (NIDS) that reframes system and network logs as natural language to detect sophisticated and previously unseen cyber threats. This presentation covers: -Limitations of traditional signature- and anomaly-based NIDS -How treating logs as language enables deep, context-aware threat detection -A walkthrough of their LLM-based detection system and its AWS deployment -Use cases, accuracy improvements, and reduced false positives -Ethical considerations including privacy, bias, over-reliance, and model misuse -Future work around real-time processing, adaptive learning, and reproducibility 🚨 Bonus: The open-source model is available on Hugging Face → 🔍 This talk is perfect for: -SOC analysts & detection engineers -AI/ML security researchers -Blue teams exploring next-gen defense systems -Anyone curious about the intersection of LLMs and cybersecurity 🎤 About Taeyang Kim: Taeyang is a decorated Machine Learning Engineer at Pattern Inc. and a master’s student at Georgia Tech. Winner of multiple hackathons and recognized as Pattern’s 2024 Employee of the Year, Taeyang blends real-world implementation with deep academic insight. He’s a rising leader in scalable AI solutions for cybersecurity and automation. 👉 Learn more about BSidesSLC: #BSidesSLC2025 #TaeyangKim #LLM #CybersecurityAI #NetworkIntrusionDetection #NIDS #AIThreatDetection #MachineLearning #HuggingFace #PatternInc #BlueTeamTools #FutureOfCybersecurity