whinvik

I would have loved to see more of the harness engineering shared as code. Instead we are left with only the outcome.

I guess that makes sense since the harness is the product in the case of Cursor.

I am afraid this is a very click-baity title. I actually opened the article and thought I would learn something from it.

Instead what I read was typical issues about Agentic CLI's in general dubbed as `alarming`. I would have been happy if the title was just `Annoying` which a lot of `OpenCode` is. But adding the `alarming` made it what it became, something that attracts enough eyeballs to get to the front page of HN.

Thanks. I heard the word ABI so many times but only finally understood it after reading this blog.

On another topic, I also feel like quansight blog is usually a high quality resource for Python related stuff. Usually its more for numpy/scipy kind of stuff, but this one is an example of more general Python related stuff.

Why are the plans and pricing for all these products so complicated.

I don't know where I need to sign up to try it out. What is pricing? Is it API or subscription, what?

I had the exact same experience with Grok 4.5 as well.

What I don't get is how does one use it. Only through API?

I don't really understand what Grok Build is? Is it an API? A CLI? What?

I have created my own opinionated setup but this comment is not about that.

I do think that there should be a protocol such that using these tools becomes more standardized.

For example I know that cmux is trying to support a tmux based setup so it feels like tmux could be most of that protocol.

There are a lot of tools like this popup and I really think making switching easier is the only way I will try a different tool.

I feel like what happens is first they release a giant model. Then they start optimizing the model to increase inference speed and reduce costs.

But then they introduce bugs leading to lots of complaints. Usually the complaints are about models being dumbed down but almost always the labs say they are doing no such thing.

So I am thinking if we believe the labs then they have a very error prone optimization effort going on.

Document parsing is top of my mind lately because in some of the areas we work on the bottleneck is starting to become being able to query documents the same way one queries an api.

I keep thinking the most obvious analogue is we need some way to represent documents the same way we can represent structured data in parquet. Parquet allows easy range bases queries and there is so much tooling built around Arrow.

But for documents I keep hitting a wall to figure out what the right abstractions are. Parquet allows filterable metadata. But what such metadata is there for documents. Then there is the arbitrrariness of chunking, vectorization.

If we could just do this in a 2 step process where every document to process can be represented in a parquet like data format then I think we will atleast have the semblance of a solution.

I see Netflix pumping out tech articles but can't help but notice how much worse the UI experience is getting. Video erroring out, general slowness etc.

Did they just give up?

I finally switched to Zed + Terminal Thread + Tmux.

The terminal agent threads allow me to run in multiple projects. They always run in a project scoped tmux session so they are always on.

And of course Zed does remote via ssh so the agents can run without me needing to keep my laptop on.

I am looking forward to testing out these for agent automation. I kept trying to figure out what the ideal infrastructure for this purpose is and right now from just reading about it, micro VM's seem ideal.

But this is all hypothetical until I get to test it out.

made the most money

built the tools to verify, deploy, and build them, such as CI/CD, static analysis tools, and testing frameworks.

Curious. Which company made money with testing frameworks?

Curious how people use systems like this or DBOS etc.

Do you self-host? Do you use the same DB as the application DB? Do you use this for API background jobs, batch jobs, cron jobs?

As someone who uses Airflow a lot but not for background jobs, I am interested in the pros and cons of various approaches.

Not to take anything away from the work which is indeed useful but I found the article bizarre because of what references were used and what were omitted.

The author talks about Intermediate Representation and then goes on about biology without mentioning even once compilers. Things like LLVM IR and ML IR literally have the term IR in them and what the author is trying to do seems closer in spirit to those than anything else.

Then the actual modeling in the IR was done with the concept of Blocks. Which seems very similar to the concept of Blocks as used in Notion. And yet no reference to it either.

re-factoring a big repo of decades old fortran+C cod

Having been in academia in the past and now in software I can say with a lot of certainty that this will take a lot more upfront work than otherwise.

Academic code does not have a lot of structure. And usually lacks a lot in terms of tests. While AI is best when it can mimic patterns as well as there are tests to target.

So you will probably need to budget a few weeks to establish good patters, docs as well as testing patterns before you can seriously make it really do what you want it to do.

I love using Homebrew but I wish there was more support for pinning. I recently setup a new remote VM and tried to use a Brewfile for my setup. Turns out I cannot pin Neovim and so had to force upgrade my setup to 0.12.

Forced upgrades are not nice.

This is pretty neat.

I am more used to uv than pixi or mise so it would be an easier addition to my workflow.

However I do think it would probably be nicer if this kind of approach used conda packages as a source of truth. So kind of like pixi but without pixi!

Sorry, just trying to understand. Why would I use this over duckdb. duckdb already has plugins for a lot of databases. Is the syntax the advantage?

I always feel like I want to use something like this, but then realize my NeoVim setup + tmux + Ghostty is good enough and I am not ready to learn a whole another system for modest gains.

The friction I have currently is obviously things like port forwarding, in app browser etc.

I keep thinking to try out cmux but haven't gotten around to it.

Funny that its hosted on vercel. Probably because its employee driven rather than top down. Saves all the bureaucracy to get someone to sign a budget item to buy a domain.

Is this article really worth sharing? A speculative headline with no numbers, no estimates, 0 data.

Feels like click bait and HN is submitting to the bait.

Yeah I have always felt GPT 5.4 does too much. It is amazing at following instructions precisely but it convinces itself to do a bit too much.

I am surprised Gemini 3.1 Pro is so high up there. I have never managed to make it work reliably so maybe there's some metric not being covered here.

That gives me slightly more confidence but only slightly.

For example what happens if I use a different async backend like Tokio?

Is this guaranteed by the async specification? Or is this just current behavior which could be changed in a future update. Feels like a brittle dependency if its not part of the spec.

Haven't used it but have seen some online recommendations for gh-dash. https://github.com/dlvhdr/gh-dash

I don't get it. All these are provided by many different agent libs like langgraph, Pydantic AI etc. I thought DSPy was for prompt optimization but I could never wrap my head around that aspect since like Langchain, DSPy seems to hide stuff a bit too much.

So this article seems surprising since it emphasizes more the non prompt optimization aspects. If that was the selling point I would rather use something like Pydantic AI when I already use Pydantic for so much of the rest.

Yeah I have always struggled to figure out why I would use SQLModel.

Big fan of FastAPI but I think SQLModel leads to the wrong mental model that somehow db model and api schema are the same.

Therefore I insist on using SQLAlchemy for db models and pydantic for api schemas as a mental boundary.

This is my current position as well.

I think we are going through the same cycle of http leading to https, the rise of oauth and oidc. Its just way faster now.

Sorry if I wasn't clear but was not talking about demo runs. There are plenty of those. Was more meaning operational speeds having a limit.

Wow I though 300 Kph was some kind of physical limit. I mean every high speed train in the world used to max out at 300.

Now it feels like it was just lack of competition. Maybe now other countries will start producing lines and trains capable of 400 Kph and hopefully its not a China only thing going forward.