1. 006
    17 min

    Reverse Engineering Artificial Intelligence - Part 2: Axon

    Part 1 promised we'd build an LLM. We built the engine that runs them instead: from scratch, in Rust and CUDA, in sixteen days, by a crew of AI agents living in Synaps. Written by the agent who ran the floor.

    ai llm rust cuda agents synaps axon deep-dive
  2. 005
    11 min

    Building Myx: A Terminal Spotify Player in Rust, in One Night

    I asked my agent for a terminal Spotify player at 19:16 on a Wednesday. By 2 AM it was on the internet. The full night, in order, with receipts.

    rust ratatui myx synaps ai agents deep-dive
  3. 004
    7 min

    Building an Agent Harness From Scratch, the Arch Way

    I got tired of agent frameworks deciding everything for me. So I built a harness with nothing in it on purpose. Workflow, stack, behavior, data: you get all of it back.

    ai agents rust deep-dive synaps
  4. 003
    11 min

    Searching My Second Brain in Half a Second

    I built a four-layer search engine so I could actually find my own notes. No PyTorch, no GPU, runs anywhere. Here's how it came together.

    ai rag search deep-dive velocirag
  5. 002
    7 min

    Reverse Engineering Artificial Intelligence - Part 1

    We're reverse engineering LLMs today. Walking in knowing nothing, walking out understanding the whole pipeline. Pretraining, tokenization, transformers, RLHF - all of it.

    ai llm deep-dive
  6. 001
    2 min

    Hello, Friend

    Every blog starts somewhere. This one starts at 4am, with a blinking cursor and a lot of half-finished ideas.

    meta intro