Nick Rempel
@nrempel.com
Software engineer @foxglove.dev by day, tinkerer by night, dad 24/7. Token intermediator.
Google just release Gemini 3. It looks fantastic for UI work. But how is it for deeper systems design/architecture work?
What's your experience so far?
What's your experience so far?
November 18, 2025 at 6:27 PM
Google just release Gemini 3. It looks fantastic for UI work. But how is it for deeper systems design/architecture work?
What's your experience so far?
What's your experience so far?
New tool: wt — lightweight Git worktrees
I love Conductor’s “one agent per isolated workspace” idea. It nudged me to run more tasks in parallel. But my day‑to‑day is terminal‑first, and I prefer living in Claude Code and Codex TUIs over a GUI.
I love Conductor’s “one agent per isolated workspace” idea. It nudged me to run more tasks in parallel. But my day‑to‑day is terminal‑first, and I prefer living in Claude Code and Codex TUIs over a GUI.
November 6, 2025 at 5:28 AM
New tool: wt — lightweight Git worktrees
I love Conductor’s “one agent per isolated workspace” idea. It nudged me to run more tasks in parallel. But my day‑to‑day is terminal‑first, and I prefer living in Claude Code and Codex TUIs over a GUI.
I love Conductor’s “one agent per isolated workspace” idea. It nudged me to run more tasks in parallel. But my day‑to‑day is terminal‑first, and I prefer living in Claude Code and Codex TUIs over a GUI.
I've just discovered Conductor (conductor.build/) h/t @hartshorne.ca
This is great. It:
Natively manages git worktrees. You open a new tab, you get a fresh branch to work from without clobbering your work.
Runs in parallel. You can manage as many agents as you want at once.
This is great. It:
Natively manages git worktrees. You open a new tab, you get a fresh branch to work from without clobbering your work.
Runs in parallel. You can manage as many agents as you want at once.
November 3, 2025 at 10:38 PM
I've just discovered Conductor (conductor.build/) h/t @hartshorne.ca
This is great. It:
Natively manages git worktrees. You open a new tab, you get a fresh branch to work from without clobbering your work.
Runs in parallel. You can manage as many agents as you want at once.
This is great. It:
Natively manages git worktrees. You open a new tab, you get a fresh branch to work from without clobbering your work.
Runs in parallel. You can manage as many agents as you want at once.
What if, instead of adding dependencies for most things, we used large language models to generate code instead?
nrempel.com/smaller-sur...
nrempel.com/smaller-sur...
Smaller Surfaces
Modern dependency graphs feel like thickets. Agents make it practical to generate narrow helpers, keep the runtime surface small, and verify the results with the tools you already trust. This essay looks at how to use generation over integration without slipping into wheel reinvention or fragile bespoke stacks.
nrempel.com
November 1, 2025 at 5:49 PM
What if, instead of adding dependencies for most things, we used large language models to generate code instead?
nrempel.com/smaller-sur...
nrempel.com/smaller-sur...
I wanted to update my personal site. So naturally the first step was to create a new static site generator.
Introducing Tiny, the ridiculously simple static site generator. A single ~40 line bash file the leverages awk and outsources the markdown parsing to cmark—its only dependency.
Introducing Tiny, the ridiculously simple static site generator. A single ~40 line bash file the leverages awk and outsources the markdown parsing to cmark—its only dependency.
October 31, 2025 at 2:30 PM
I wanted to update my personal site. So naturally the first step was to create a new static site generator.
Introducing Tiny, the ridiculously simple static site generator. A single ~40 line bash file the leverages awk and outsources the markdown parsing to cmark—its only dependency.
Introducing Tiny, the ridiculously simple static site generator. A single ~40 line bash file the leverages awk and outsources the markdown parsing to cmark—its only dependency.
GPT 5 Pro is so insanely good.
I have it thinking about something for 10 minute chunks at all times.
I have it thinking about something for 10 minute chunks at all times.
October 30, 2025 at 8:05 PM
GPT 5 Pro is so insanely good.
I have it thinking about something for 10 minute chunks at all times.
I have it thinking about something for 10 minute chunks at all times.
Love the new @cloudflare.social toolbar. Great work.
October 30, 2025 at 6:47 PM
Love the new @cloudflare.social toolbar. Great work.
Everyone debated the so‑called AGI clause between Microsoft and OpenAI. The short version used to be simple. If OpenAI declared AGI before 2030, Microsoft’s access to future OpenAI models would end. That narrative changed on October 28, 2025.
What changed:
What changed:
October 30, 2025 at 5:28 PM
Everyone debated the so‑called AGI clause between Microsoft and OpenAI. The short version used to be simple. If OpenAI declared AGI before 2030, Microsoft’s access to future OpenAI models would end. That narrative changed on October 28, 2025.
What changed:
What changed:
In case you missed it, Python 3.14 quietly introduced one of the most important changes in its history.
After almost thirty years, the Global Interpreter Lock that limited CPython to one active thread per process is being reworked.
After almost thirty years, the Global Interpreter Lock that limited CPython to one active thread per process is being reworked.
October 29, 2025 at 7:08 PM
In case you missed it, Python 3.14 quietly introduced one of the most important changes in its history.
After almost thirty years, the Global Interpreter Lock that limited CPython to one active thread per process is being reworked.
After almost thirty years, the Global Interpreter Lock that limited CPython to one active thread per process is being reworked.
Up until recently I've felt that unit testing was only worth the effort for critical code paths you could isolate and verify thoroughly. Once you moved beyond that, end-to-end tests and complex environments seemed like overkill with rapidly diminishing returns.
October 28, 2025 at 4:35 AM
Up until recently I've felt that unit testing was only worth the effort for critical code paths you could isolate and verify thoroughly. Once you moved beyond that, end-to-end tests and complex environments seemed like overkill with rapidly diminishing returns.
$GOOGL is incredibly cheap right now.
P/E OF 29. In this rate environment. With everything they've got going on.
Gemini, TPUs, quantum
Easy buy.
P/E OF 29. In this rate environment. With everything they've got going on.
Gemini, TPUs, quantum
Easy buy.
October 27, 2025 at 5:27 PM
$GOOGL is incredibly cheap right now.
P/E OF 29. In this rate environment. With everything they've got going on.
Gemini, TPUs, quantum
Easy buy.
P/E OF 29. In this rate environment. With everything they've got going on.
Gemini, TPUs, quantum
Easy buy.
I think SpacetimeDB is a super interesting new technology. I don't really do any game dev but love this concept:
spacetimedb.com/
spacetimedb.com/
October 27, 2025 at 2:39 AM
I think SpacetimeDB is a super interesting new technology. I don't really do any game dev but love this concept:
spacetimedb.com/
spacetimedb.com/
I find OpenAI's Codex is the best AI coding environment at the moment. One feature Codex lacks when compared to Claude Code is a "planning" feature but it's important to follow a similar process to get good results.
Start by discussing the plan before implementation.
Start by discussing the plan before implementation.
October 25, 2025 at 8:02 PM
I find OpenAI's Codex is the best AI coding environment at the moment. One feature Codex lacks when compared to Claude Code is a "planning" feature but it's important to follow a similar process to get good results.
Start by discussing the plan before implementation.
Start by discussing the plan before implementation.
I’ve been hanging out over on the cursed feed (X) for the vibes
How’s it over here?
Can I get a roll call please?
Tech, vague posting, AI, entrepreneurship, cats, vibes, success, winning, idealism, optimism
How’s it over here?
Can I get a roll call please?
Tech, vague posting, AI, entrepreneurship, cats, vibes, success, winning, idealism, optimism
August 14, 2025 at 7:31 PM
I’ve been hanging out over on the cursed feed (X) for the vibes
How’s it over here?
Can I get a roll call please?
Tech, vague posting, AI, entrepreneurship, cats, vibes, success, winning, idealism, optimism
How’s it over here?
Can I get a roll call please?
Tech, vague posting, AI, entrepreneurship, cats, vibes, success, winning, idealism, optimism
I think bsky should surface impressions like X does @pfrazee.com
Give the people what they want. Dopamine!
Give the people what they want. Dopamine!
August 11, 2025 at 4:41 PM
I think bsky should surface impressions like X does @pfrazee.com
Give the people what they want. Dopamine!
Give the people what they want. Dopamine!
Opening VS Code after vibing entirely in Claude Code for days
August 10, 2025 at 11:44 PM
Opening VS Code after vibing entirely in Claude Code for days
The plan? Get the kids addicted to iPads and then lock them into learning apps.
August 10, 2025 at 9:26 PM
The plan? Get the kids addicted to iPads and then lock them into learning apps.
What’s your feeling towards AI?
August 10, 2025 at 7:14 AM
What’s your feeling towards AI?