// BLOG · DEEP DIVES
The things most people
hand-wave past.
Investigations into how systems actually work, written while I was breaking them. No tutorials copied off the docs — these are the notes from tracing, profiling, and proving it.
ALL POSTS · 92
AIAn AI Coworker in an LXC ContainerJuly 12, 20264 minNETWORKINGWho's Allowed to Talk to What: Identity and Access in the HomelabJuly 12, 20266 minLINUXThe Fleet: Six Machines, and What the Hypervisor Can't DoJuly 12, 20267 minEMBEDDEDWaking the Lab: an ESP32, MQTT, and 102 BytesJuly 12, 20266 minLINUXMy Homelab: One Proxmox Box, ZFS Pools, and a Properly Segmented NetworkJuly 10, 20268 minWEBHow This Blog Runs for $0 a MonthJuly 8, 20266 minAIHow LLM Inference Actually Runs: Tokens, KV Cache, and Why It's Memory-BoundJuly 5, 20265 minPERFORMANCEReading Flame Graphs: Where Your CPU Time Actually GoesJuly 1, 20265 minLINUXmmap vs read(): When Mapping a File Actually WinsJune 22, 20263 minKERNELAnatomy of a Context Switch: What the Kernel Does Between Two ThreadsJune 20, 20263 minFUNDAMENTALSFloating Point: Why 0.1 + 0.2 Isn't 0.3June 15, 20263 minALGORITHMSBloom Filters: Probably Present, Definitely UsefulJune 12, 20263 minGITGit Internals: Your Repo Is a Content-Addressed DatabaseJune 10, 20263 minOBSERVABILITYeBPF: Running Your Code in the Kernel Without a ModuleJune 4, 20263 minCONTAINERSWhat a Container Actually Is: Namespaces and cgroupsJune 2, 20263 min