Posts
All posts from William Zujkowski - security insights, AI/ML projects, and career development
Building My Digital Garden with Eleventy
Why I chose Eleventy for my personal site and the journey of building a fast, accessible, privacy-respecting digital home with modern web technologies
GPU Power Monitoring in My Homelab: When Machine Learning Met My Electricity Bill
Monitoring GPU power consumption during ML training with NVIDIA SMI, custom dashboards, and optimization strategies to reduce electricity costs
Blockchain Beyond Cryptocurrency: Building the Trust Layer of the Internet
Running Ethereum nodes on my homelab taught me blockchain is about building trust without central authorities. Here's what works (and doesn't)
Sandboxing Untrusted Containers with gVisor: Lessons from G-Fuzz Vulnerability Research
Deploying gVisor for application-level container sandboxing in my homelab K3s cluster, protecting against kernel exploits while managing the 15-30% performance overhead.
Running LLaMA 3.1 on a Raspberry Pi: Memory-Efficient Edge AI with PIPELOAD
How I deployed a 7B parameter language model on an 8GB Raspberry Pi using PIPELOAD's pipeline inference: 90% memory reduction, 4x faster inference.
Quantum Computing's Leap Forward
Recent quantum computing breakthroughs bring us closer to practical quantum advantage—algorithm development, error correction, and real applications
Multimodal Foundation Models: Capabilities, Challenges, and Applications
Foundation models that understand text, images, and audio together—architecture, capabilities, and applications beyond single-modality systems
Sustainable Computing: Strategies for Reducing IT's Carbon Footprint
ML training consuming as much electricity as a small town sparked my journey into sustainable computing—efficiency as environmental imperative
Zero Trust Architecture: A Practical Implementation Guide
Practical guide to implementing zero trust architecture: identity verification, micro-segmentation, and continuous monitoring for modern security
Designing Resilient Systems for an Uncertain World
Building systems that gracefully handle failures through redundancy, circuit breakers, and chaos engineering—lessons from production incidents