I Fine-Tuned a Security Reasoning Model That Runs on a 4GB Laptop (No GPU, No Cloud)

How I fine-tuned DeepSeek-R1-Distill into a cybersecurity SLM covering 2026 AI-native attacks — MCP poisoning, agentic lateral movement, Crescendo jailbreaks — and quantized it to 1.2GB GGUF using free Colab.