How AI inference works, from the transistor to the token, plus chip design and building with agents.https://juansync7.github.io/https://juansync7.github.io/learn/agents-you-can-repeat/https://juansync7.github.io/learn/agents-you-can-repeat/The agents I trust aren't the clever ones. They're the ones whose every decision a script could make has been taken away from them.Sat, 24 Oct 2026 00:00:00 GMThttps://juansync7.github.io/learn/a-clock-target-is-a-budget/https://juansync7.github.io/learn/a-clock-target-is-a-budget/"Run at 2.6 GHz" sounds like one number, but it's a budget of about 385 picoseconds that four different teams have to share.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/build-the-bump-not-the-sign/https://juansync7.github.io/blog/build-the-bump-not-the-sign/Most of the rules I wrote for my agents either never reached them or quietly stopped being true, until I started writing the check instead.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/chip-habits-in-software/https://juansync7.github.io/blog/chip-habits-in-software/The habits I brought from chip design mostly help, and the few that lie look exactly like the ones that don't.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/decide-slowly-move-fast/https://juansync7.github.io/blog/decide-slowly-move-fast/Before you build is the cheapest window a project will ever have, and it's the one we rush.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/second-brain-from-my-chats/https://juansync7.github.io/blog/second-brain-from-my-chats/I couldn't remember half of what I'd built, so I had my AI chats write it down, and the hard part was keeping the record honest.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/teaching-an-ai-my-voice/https://juansync7.github.io/blog/teaching-an-ai-my-voice/I gave an AI a careful list of rules for how I write, and every draft followed them and still didn't sound like me.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/unknown-is-not-zero/https://juansync7.github.io/blog/unknown-is-not-zero/A missing value that turns into a zero looks like an answer, and that is exactly why it gets trusted.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/what-a-hash-proves/https://juansync7.github.io/blog/what-a-hash-proves/A hash proves the bytes haven't changed, which is a lot, and it's very easy to let it vouch for much more than that.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/learn/where-the-gpu-memory-goes/https://juansync7.github.io/learn/where-the-gpu-memory-goes/On a shared GPU, the memory for every request is set aside before the first one arrives, and most serving surprises come from not knowing where it went.Sun, 18 Oct 2026 00:00:00 GMThttps://juansync7.github.io/blog/how-i-draw-figures/https://juansync7.github.io/blog/how-i-draw-figures/Every drawing here is a small file run by one clock and checked by an audit that photographs it. Some of those files are templates: a seed, seeded noise and a grammar, and a different picture for every number.Sat, 17 Oct 2026 00:00:00 GMThttps://juansync7.github.io/learn/inference-end-to-end/https://juansync7.github.io/learn/inference-end-to-end/A visual walkthrough of LLM inference from prompt to token — prefill vs decode, the KV-cache, precision, speculative decoding, the memory hierarchy, and how the big labs serve at scale.Sun, 14 Jun 2026 00:00:00 GMThttps://juansync7.github.io/learn/how-inference-chips-are-built/https://juansync7.github.io/learn/how-inference-chips-are-built/From transistors and dataflow to wafer-scale engines — a visual tour of how the silicon behind AI inference is designed, and why startups think they can unseat the GPU.Fri, 12 Jun 2026 00:00:00 GMThttps://juansync7.github.io/blog/rtl-to-gds/https://juansync7.github.io/blog/rtl-to-gds/From Verilog to tapeout — the full ASIC design flow broken down into digestible steps. Covering synthesis, P&R, timing closure, and the dark arts of DRC.Tue, 28 Apr 2026 00:00:00 GMThttps://juansync7.github.io/blog/cyberpunk-web/https://juansync7.github.io/blog/cyberpunk-web/How I turned terminal aesthetics and chip schematics into a personal site. Starfields, FSM diagrams, and unhealthy amounts of CSS.Wed, 15 Apr 2026 00:00:00 GMThttps://juansync7.github.io/blog/fpga-neural/https://juansync7.github.io/blog/fpga-neural/Deploying quantized models on Xilinx Zynq. Covers HLS, resource budgeting, and why your first attempt will always be too slow.Mon, 30 Mar 2026 00:00:00 GMThttps://juansync7.github.io/blog/vim-asic/https://juansync7.github.io/blog/vim-asic/Custom keybinds, SystemVerilog LSP, waveform previews, and why I refuse to use an IDE.Wed, 18 Mar 2026 00:00:00 GMThttps://juansync7.github.io/blog/timing-closure/https://juansync7.github.io/blog/timing-closure/When your critical path laughs at your clock constraint. Strategies, hacks, and copium for meeting timing.Thu, 05 Mar 2026 00:00:00 GMThttps://juansync7.github.io/blog/react-terminal/https://juansync7.github.io/blog/react-terminal/State machines, command parsing, and typing animations. How the work experience section actually works under the hood.Fri, 20 Feb 2026 00:00:00 GMThttps://juansync7.github.io/blog/silicon-vs-software/https://juansync7.github.io/blog/silicon-vs-software/Two worlds, same debugging despair. A comparison of methodologies, timelines, and the meaning of "shipping."Sun, 08 Feb 2026 00:00:00 GMThttps://juansync7.github.io/blog/starfield-canvas/https://juansync7.github.io/blog/starfield-canvas/Twinkling stars, shooting stars, parallax layers — building an ambient background that doesn't murder your GPU.Sun, 25 Jan 2026 00:00:00 GMThttps://juansync7.github.io/blog/aion-chip/https://juansync7.github.io/blog/aion-chip/A non-NDA-violating look at life as a Senior ASIC Design Engineer. Spoiler: lots of waveform staring.Sat, 10 Jan 2026 00:00:00 GMThttps://juansync7.github.io/blog/risc-v-core/https://juansync7.github.io/blog/risc-v-core/A weekend project that took three months. 5-stage pipeline, hazard detection, and the satisfaction of running your own ISA.Sat, 20 Dec 2025 00:00:00 GMThttps://juansync7.github.io/blog/python-eda/https://juansync7.github.io/blog/python-eda/TCL is pain. Here's how I replaced 2000 lines of TCL with 200 lines of Python and a lot of subprocess calls.Fri, 05 Dec 2025 00:00:00 GMThttps://juansync7.github.io/blog/coffee-ranking/https://juansync7.github.io/blog/coffee-ranking/An empirical study. N=23. Methodology: caffeine addiction. Results: inconclusive but delicious.Tue, 18 Nov 2025 00:00:00 GMT