← Feed

Posts

• Post #3006805

  AI agents aren't magic, they are LLMs wrapped in a "harness" of code and bring both considerable benefits and risks. Step-wise building a 50-line Python script, this post demystifies how agentic loops work. The second half highlights the engineering reality behind the hype, and the hidden technical and ethical risks. https://marvin.damschen.net/post/llm-based-agents/ @localllama #llm #ai #cybersecurity

• Post #3006804

  While the average number of daily reported vulnerabilities is increasing, the Time-to-Exploit to exploit for confirmed exploited vulnerabilities is rapidly decreasing. It is clear that something needs to change, and personally, I see the EU's Cyber Resilience Act as a leap into the right direction. My take on this: https://marvin.damschen.net/post/why-cra-matters-to-you/ #cybersecurity #eu #cra

• Post #1099314

  I wrote about running LLMs locally on the Intel Arc Pro B60 GPU previously, where I used Intel's official software stack (llm-scaler / vLLM). This time I focus on the impactful open-source project llama.cpp: https://marvin.damschen.net/post/intel-arc-llama.cpp/ Taking this opportunity to test federation with lemmy 😊: @localllama #llm #selfhosting #privacy #intel #llamacpp