Nicolas Marchildon

Software developer building tools to better leverage AI | The Mobility House, CWP Energy, Plusgrade, Hopper, Intelerad

[Nicolas Marchildon @NicoMarchildon]

@l3m0v1c1 @pidotdev The ~/.pi/agent/prompts/ directory is the global location for prompt templates. Markdown files placed here are automatically loaded and can be invoked by typing /filename (without the extension). Templates support bash-style arguments such as $1, $2, $@, and ${@:N:L}

[Nicolas Marchildon @NicoMarchildon]

@MrApeXAcademy @infmes @OpenRouter FYI @OpenRouter

[Nicolas Marchildon @NicoMarchildon]

China does not have that yet, but when they do, they’d rather have a world of open models.

[Nicolas Marchildon @NicoMarchildon]

Reality is open. « Closed » models is a fool’s errand.

The moat is the inference capacity.

[Dean W. Ball @deanwball]

Some observations on Kimi:

1. It's a very good model! I don't think its performance can be explained away by distillation or anything like that. In agentic coding sessions, it seems pretty much on par with the best public models of Q1 2026. In my fairly limited use, it also seemed very token hungry. It's not obvious to me that this model is actually that cheap to run.

2. I am personally surprised the Chinese state continues to allow the open sourcing of models this good, given potential risks. To be clear, I *myself* might be fine with models presenting this level of marginal risk being open weight, but I am surprised that China is fine with it. I suspect the reason they are is 75% explained by strategic blindness/lack of AGI-pilledness (the CCP is very Yann Lecun-y in its views of AI). The other 25% or so is their lack of compute for customer inference (making China's open-weight strategy an unintended byproduct of US export controls) and the normal Chinese strategy of aggressive exports. For the companies, as opposed to the government, the decision to open source is partially ideological and partially because they are behind, and they know that very few people would pay for sub-frontier models from China.

3. Open-weight models are inherently decelerationist, and I'm continually surprised to see the so-called "accelerationists" so excited about open-weight models. I suspect the reason they are is that they know open-weight models are effectively ungovernable, and they simply like the overall cloak of ungovernability open-weight models create over the whole of AI. It's not a bad strategy; it reminds me of James Scott's recounting of the hill people in "the art of not being governed." Still, in the end, open-weight models deter further AI capex.

4. One probable outcome of an open-weight-model-dominant world is full AI communism, which is precisely what China proposes: rather than a market product, AI is a "public good" which will ultimately be provided by the state as a kind of "digital public infrastructure." This future strikes me as a dystopian hellscape, but I've never met an open-weight models advocate who doesn't ultimately concede this is where things end. You'd be surprised how many 'accelerationists' lobbied me, while I was in government, to support an eleven or twelve-figure federally funded data center so that startups could train models at a subsidy and then give them away for free. There was no other way for AI to progress, they said. Perhaps this is the logical end state of things. Nonetheless, I find myself surprised to see supposed accelerationists excited about such an outcome. I think many of them just don't know what they're doing. Many accelerationists do not view the creation and serving of frontier models as a legitimate business.

5. I would guess that the Trump Administration will at some point realize that their best strategy here would be to create large amounts of regulatory risk around the use of open-weight Chinese models. You don't need to "ban open source" (one of the dumber motifs of AI policy discussion). You just need to direct every agency to issue soft law that creates FUD. "A Federal Reserve Advisory Bulletin found that there may be backdoors in Chinese AI models." It needn't be that well justified. You just create enough regulatory risk that every regulated enterprise backs off. You probably don't want to create so much regulatory risk that you scare off the hyperscalers from serving Chinese models; this will just drive startups to sketchier providers. There's a happy middle ground here. I'd assume they will do some version of this.

6. It's probably true that open-weight models of this capability make the world a bit more dangerous, but not so much more that you'll really notice. At some point the models will be capable enough that you will notice. "A nonliving, invisible, dangerous, and infinitely self-replicating agent escaped from a Chinese lab," you say? Color me shocked.

[Nicolas Marchildon @NicoMarchildon]

Exciting times!

[Amjad Masad @amasad]

https://t.co/9ujCmLs3lQ

[Nicolas Marchildon @NicoMarchildon]

Insightful.

[Vikram Aditya @viks_rum]

https://t.co/rlFVGyXeCo

[Prizmal @PrizmalAi]

This Week in Inference: Ollama hit 8.9M monthly users and just raised $65M to keep running models locally and in the cloud. Local inference is not a niche anymore. Plus 3 more stories. Thread

[Nicolas Marchildon @NicoMarchildon]

@jasonzhou1993 Are you looking for pull requests?

[Nicolas Marchildon @NicoMarchildon]

@jasonzhou1993 Still not clear

[Nicolas Marchildon @NicoMarchildon]

@alexatallah Please tell us more!

[Nicolas Marchildon @NicoMarchildon]

I have now fully switched to Pi.

No longer hand holding Windsurf, but still involved in reviewing pull requests made by the agent. Using ghpr-monitor is how I interact with the agent 80% of the time. I simply add review comments, and it works on it while I do other stuff.

[Nicolas Marchildon @NicoMarchildon]

I have been using @windsurf as my main driver for a while. I review every diff, select text to ask AI to tweak it, etc. I'm very involved when reviewing a README, for example.

Using pi coding agent, I realize how difficult this is. It's probably why I see readme SLOP everywhere.

[Nicolas Marchildon @NicoMarchildon]

@caspar_br @grok what are the UX patterns in https://t.co/yrFJWzC7i2 that stand out?

[Nicolas Marchildon @NicoMarchildon]

https://t.co/4fqiaSxJpN

[Nicolas Marchildon @NicoMarchildon]

@Awe_automation @OpenRouter GPU running hot? I hope you have access to an electrical outlet, just just this USB port!

[Nicolas Marchildon @NicoMarchildon]

@OpenRouter @paradite_ Can you break it down by app?

[Nicolas Marchildon @NicoMarchildon]

@MiniMax_AI @kilocode "Me using Claude Opus to rename a file"
https://t.co/bukyUmEY3Z

[Nicolas Marchildon @NicoMarchildon]

@jarredsumner Done. Added await command to the unmaintained gh-pr-review extension: https://t.co/OvwsAUeK5y

Feedback/pull request welcome.

[Nicolas Marchildon @NicoMarchildon]

Introducing `gh pr-review await`: polls a PR until it needs attention — new comments, conflicts, or CI failures — then exits with a non-zero code so scripts and CI can react immediately.

Built into the gh-pr-review (https://t.co/BEPtUpcT6z) extension (a maintained fork of agynio/gh-pr-review that merged outstanding upstream PRs left unmerged for months).

Unlike my pi-ghpr-monitor, it can't wake an idle agent or inject notifications mid-session — it's just a CLI command. But for anyone scripting against GitHub PRs without Pi, it's the simplest way to wait until there's something to do.

gh pr-review await -R owner/repo 42

gh extension install elecnix/gh-pr-review