Viewing profile — mrothroc

[flagged]

Relevant passage from Jarred's post: "At the time of writing, about 4% of Bun's Rust code sits inside an unsafe block (~13,000 unsafe keywords across ~27,000 lines / ~780,000 lines…

[dead]

Drilling into the original article where Jarred explained the reasoning behind the change, It's pretty clear that under zig the team was doing things by hand that are automatic in …

Agents are fast and powerful, and that is a double-edged sword. The volume you can produce is breathtaking, but they make far too much for you to review every line of output. Perso…

You have to have a multi-layered approach. This is the "Swiss cheese" model of prevention: individual layers may have holes, but if none of the holes in the stack line up, nothing …

I break this up into two parts: assets and workflow state. All assets go in the repo, this is the context that all agents read to be able to understand what we're doing and how we …

The biggest strength of "would this get merged" is that it is actually a compound property: does it work, does it match convention, is it maintainable. And of course: does the revi…

Yes, at some point AI will be fully integrated into society so that there are entire autonomous sections. But society doesn't move at the same pace as technology. The output of age…

Let me ask you this: would people accept a nuclear power plant whose safety control software is vibe coded? Would a CFO of a public company sign off on financial statements created…

The easy decision is to just go with the biggest SOTA model you can afford. But this overlooks the other critical part of getting the most out of these things: the harness. I run a…

Some of it can be busywork, but for me the intermediate artifacts (plans, design docs, etc) serve a real purpose: they create a verification surface where you can check that the ag…

Thanks, glad you find it useful! Feel free to ping me if you have any questions.

I've been specializing in distributed systems for nearly 35 years. I've read your work, and it's shaped my thinking. When you say you have a person in mind when you write, I am tha…

Definitely stacks. The thing that made it clear for me was being explicit about the stages, and where/what you can verify with a guardrail, or gate. I wrote up the framework I use …

Yes, "guardrails" is a squishy term. But it gets clearer if you ask what transition is being guarded. Some of this is inside the model, like topic refusals. Forge sits at the tool …

I fully agree with the idea: above a model capability threshold, the power comes from the harness far more than the model. Engineers can get tremendous power from learning how to d…

I have the same experience. I've been running sequential agents in my own harness that is a standard SDLC pipeline (plan, design, code, build, test). It has gates between each stag…

The "blurring" framing makes Simon's tension sound intrinsic when it is actually structural. Vibe coding and agentic engineering aren't on a continuum. They're distinguished by the…

The list in the article looks like verification practices. Document intent, develop taste, find the hard stuff, etc. It assumes that when code is cheap the bottleneck shifts to kno…

From a verification-topology angle, what makes algotune.io contamination-resistant? Is it because the correctness oracle is a performance metric (which can't be memorized) rather t…

Simple example to show how configs are defined: { "name": "plain_3L", // Minimal causal transformer baseline: 3 attention layers plus 3 SwiGLU layers. "model_dim": 128, "vocab_size…

I built a tool for quickly testing different ML architectures. Define a model in JSON, train on your Mac (Metal) or ship the same config to a cloud GPU (CUDA). No code changes betw…

I addressed this in my reply to kelseyfrog above. The short version: the production work is proprietary, the tooling I used to do the analysis is open source.

Hi, I'm the original author and I can clarify a few things. The 543 hours are the agent compute hours, not me at the keyboard. The pipeline runs autonomously, the agents execute in…