Hacker Newsnew | past | comments | ask | show | jobs | submit | f311a's commentslogin

You don't want a browser with a bunch of RCEs that can be triggered by opening a web page...

You do want a browser with RCE, but you want it to keep the it sandboxed. The hard part is executing the code safely

It's good enough if you don't go wild and allow LLMs to produce 5k+ lines in one session.

In a lot of industries, you can't afford this anyway, since all code has to be carefully reviewed. A lot of models are great when you do isolated changes with 100-1000 lines.

Sometimes it's okay to ship a lot of code from LLMs, especially for the frontend. But, there are a lot of companies and tasks where backend bugs cost a lot, either in big customers or direct money. No model will allow you to go wild in this case.


You are losing your personality by modifying your text with LLMs. It saves you how much, 1 minute of writing?

> isn't just "high demand", but "contractual lock-out."

The "isn't just .., but .." construction is so overused by LLMs.


You are living in a fantasy. We are nowhere near this, and engineers who understand fundamentals very well are the ones in demand.

We will see, I still think we will need engineers who understand fundamentals, and yes we need to teach new devs these like we teach them how assembly, computer architecture works, even though we likely don't build our own compilers or hardware. But as time goes on there is less need to know about the lower levels.

It's already, that's why Microsoft abandoned GitHub


That's because they still struggle hard with out-of-distribution tasks even though some of them can be solved using existing training data pretty well. Focusing on out-of-distribution will probably lower scores for benchmarks. They focus too much on common tasks.

And keep in mind, the original creators of the first compiler had to come up with everything: lexical analysis -> parsing -> IR -> codegen -> optimization. LLMs are not yet capable of producing a lot of novelty. There are many areas in compilers that can be optimized right now, but LLMs can't help with that.


They are trained pretty hard to transpile the code between languages and do this pretty well because this can be done using RL.

You can force the agent not to use unsafe, this is why it burned $20000. Thousands of attempts against good tests with good boundaries set.


No, Apple ecosystem is bad enough already in software terms. Just let me use my computer as I want.

"An idiot admires complexity, a genius admires simplicity." Terry A. Davis


Just look at the closed PRs of their project. General technical knowledge is so low it's insane. It attracts weird people.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: