Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A Research Preview of Codex

openai.com

Post reply on HN
Print view View original

351–360 of 487 posts

Re: A Research Preview of Codex

#351

Post by rtpg »

sam0x17 wrote: post #171
criddell wrote: post #112

Earlier quoted context omitted.

If you aren't hiring junior engineers to do these kinds of things, where do you think the senior engineers you need in the future will come from? My kid recently graduated from a very good school with a degree in computer science and what she's told me about the job market is scary. It seems that, relatively speaking, there's a lot of postings for senior engineers and very little for new grads. My employer has hired…

Hiring of juniors is basically dead these days and it has been like this for about 10 years and I hate it. I remember when I was a junior in 2014 there were actually startups who would hire cohorts of juniors (like 10 at a time, fresh out of CS degree sort of folks with almost no applied coding experience) and then train them up to senior for a few years, and then a small number will stay and the rest will go elsewhe…

I feel like hiring juniors still exists, because I still hear about loads of "boring" small startups that do the "we hire juniors and seniors".

Juniors cuz they're cheap and motivated (and potentially grow into very good team members!), seniors to handle the trickier stuff. Mid-level people don't get picked up cuz there's enough junior pipeline that the cost-benefit doesn't work out.

Thing is these companies tend to have, say, college pipelines and the like for juniors. Internships and the like. It would be really painful to not have internships lined up at "your local startup" in this day and age.

My impression is that a lot of the junior dev pipeline is in smaller places that don't show up in job boards in the same way as the rest of it. Maybe that's dried up too, but I have my doubts.

You still need somebody to work the robot, even if the robot is "doing the coding"!

Re: A Research Preview of Codex

#352

Post by popcorncowboy »

SubiculumCode wrote: post #259
hintymad wrote: post #179

Earlier quoted context omitted.

It looks we are in this interesting cycle: millions of engineers contribute to open-source on github. The best of our minds use the code to develop powerful models to replace exactly these engineers. In fact, the more code a group contributes to github, the easier it is for the companies to replace this group. Case in point, frontend engineers are impacted most so far. Does this mean people will be less incentivized…

Now do open science. More generally, specialty knowledge is valuable. From now on, all employees will be monitored in order to replace them.

> From now on, all employees will be monitored in order to replace them.

This is going on a t-shirt.

Re: A Research Preview of Codex

#353

Post by inkyoto »

criddell wrote: post #112
johnjwang wrote: post #80

Some engineers on my team at Assembled and I have been a part of the alpha test of Codex, and I'll say it's been quite impressive. We’ve long used local agents like Cursor and Claude Code, so we didn’t expect too much. But Codex shines in a few areas: Parallel task execution: You can batch dozens of small edits (refactors, tests, boilerplate) and run them concurrently without context juggling. It's super nice to run…

If you aren't hiring junior engineers to do these kinds of things, where do you think the senior engineers you need in the future will come from? My kid recently graduated from a very good school with a degree in computer science and what she's told me about the job market is scary. It seems that, relatively speaking, there's a lot of postings for senior engineers and very little for new grads. My employer has hired…

> My employer has hired recently and the flood of resumes after posting for a relatively low level position was nuts.

I am utterly perplexed with the current situation on the job market, which seems to be a global phenomenon that is not constrained to a particular country or region. Late last year, I was hiring for two junior software engineering positions and (through an external recruitment partner) we received over 400 job applications for two junior positions. We, however, scrambled to narrow the number of candidates down to ten, out of which eight turned out to be lemons and two ended up being exceptionally good. 390 other applicants ended up being pure white noise.

Colleagues in a neighbouring business unit reported receiving over 600 submissions for a single position.

I have approached a few headhunters in the last couple of months with informal questions about what has been happening. They are under constant duress, receiving hundreds upon hundreds of applications for pretty much any position. The feedback is that when most people see a job ad, they put their resume through GenAI and submit whatever garbage comes out of it without even looking at the output. The vast majority of people can't even be bothered to write a simple cover letter, which could have been used as a shibboleth for the hiring manager / recruiter: «I am an intelligent human being, and I am real».

Naturally, the headhunters have responded with GenAI-assisted tools to sift through piles of putrid trash. The side effect is that such good, qualified applicants do not usually get a chance to get screened in.

The situation does not seem to be changing, and the only way out seems to be applying through a professional network or connections. People abusing GenAI are hurting themselves (ironically, GenAI has become pretty good at recognising GenAI-generated content), and they are also hurting pretty much everyone else in the process, and they do not care.

Re: A Research Preview of Codex

#355

Post by meta_ai_x »

criddell wrote: post #112
johnjwang wrote: post #80

Some engineers on my team at Assembled and I have been a part of the alpha test of Codex, and I'll say it's been quite impressive. We’ve long used local agents like Cursor and Claude Code, so we didn’t expect too much. But Codex shines in a few areas: Parallel task execution: You can batch dozens of small edits (refactors, tests, boilerplate) and run them concurrently without context juggling. It's super nice to run…

If you aren't hiring junior engineers to do these kinds of things, where do you think the senior engineers you need in the future will come from? My kid recently graduated from a very good school with a degree in computer science and what she's told me about the job market is scary. It seems that, relatively speaking, there's a lot of postings for senior engineers and very little for new grads. My employer has hired…

A junior engineer can always become a senior engineer by using Gemini/ChatGPT to build full systems and literally asking chatGPT, a series of Why questions to every output.

In fact not being bottle necked by senior engineers or not having to drawe the luck of a bad senior engineer/mentor, there will be a new stars of Junior engineers.

What you should be worried is Senior Engineers who hate AI

Re: A Research Preview of Codex

#356

Post by TheHegemon »

HappMacDonald wrote: post #150
bmcahren wrote: post #141

Earlier quoted context omitted.

Counter-point A: AI coding assistance tools are rapidly advancing at a clip that is inarguably faster than humans. Counter-point B: AI does not get tired, does not need space, does not need catering to their experience. AI is fine being interrupted and redirected. AI is fine spending two days on something that gets overwritten and thrown away (no morale loss).

Counter-counter-point A: If I work with a human Junior and they make an error or I familiarize them with any quirk of our workflow, and I correct them, they will recall that correction moving forward. An AI assistant either will not remember 5 minutes later (in a different prompt on a related project) and repeat the mistake, or I'll have to take the extra time to code some reminder into the system prompt for every pr…

> If I work with a human Junior and they make an error or I familiarize them with any quirk of our workflow, and I correct them, they will recall that correction moving forward

I really wish that were the case. Most of the Jr Engineers I work with have to be told the same thing multiple times, in different ways, for things to stick.

Re: A Research Preview of Codex

#357

Post by solresol »

I'm not sure what's wrong with me, but I just wasted several hours wrestling codex to make it behave.

Here's my workflow that keeps failing: - it writes some code. It looks good a first glance - I push it to github - automated tests on github show that there's a problem - go back to codex and ask it to fix it - it does stuff. It looks good again.

Now what do I do? If I ask it to push again to github, then it will often create a pull request that doesn't include stuff from the first pull request, but it's not a pull request that stacks on top of the previous pull request, it's a pull request that stacks on top of main.

When asked to write something that called out to gpt-4.1-mini, it used openai.ChatCompletion.create (!?!!?)

I just found myself using claude to fix codex's mistakes.

Re: A Research Preview of Codex

#358

Post by CuriouslyC »

Buttons840 wrote: post #329
jackphilson wrote: post #318

Earlier quoted context omitted.

Yes. everyone will eventually have the job title of "problem solver"

Don't forget the very important role of managing the problem solvers--if you just let the problem solvers run amuck all sorts of problems might be solved.

Major Dilbert vibes

Re: A Research Preview of Codex

#359

Post by fragmede »

sam0x17 wrote: post #171
criddell wrote: post #112

Earlier quoted context omitted.

If you aren't hiring junior engineers to do these kinds of things, where do you think the senior engineers you need in the future will come from? My kid recently graduated from a very good school with a degree in computer science and what she's told me about the job market is scary. It seems that, relatively speaking, there's a lot of postings for senior engineers and very little for new grads. My employer has hired…

Hiring of juniors is basically dead these days and it has been like this for about 10 years and I hate it. I remember when I was a junior in 2014 there were actually startups who would hire cohorts of juniors (like 10 at a time, fresh out of CS degree sort of folks with almost no applied coding experience) and then train them up to senior for a few years, and then a small number will stay and the rest will go elsewhe…

The problem is getting hired. With all the resources available today, learning programming is easy compared to pre-LLM, pre-Stack Overflow, pre-Google days of learning to program. I dare say an autodidact in the original dot com boom, transported to today, would be fine, as far as being useful to a company goes. You don't need to know every frontend framework, all possible backends, and be a Linux god at devops, all at once. Sure there's more stuff today then in the 00's, but no team is using all of all three of those simultaneously, so what you practically have to know isn't too much for a motivated individual to learn.

The problem is getting hired. If seniors are having problems getting callbacks for interviews right now, then a young kid with a thin resume isn't going to get hired, no matter how senior their skills are in reality.

Re: A Research Preview of Codex

#360

Post by klabb3 »

nadis wrote: post #127

In the preview video, I appreciated Katy Shi's comment on "I think this is a reflection of where engineering work has moved over the past where a lot of my time now is spent reviewing code rather than writing it." Preview video from Open AI: https://www.youtube.com/watch?v=hhdpnbfH6NU&t=878s As I think about what "AI-native" or just the future of building software loos like, its interesting to me that - right now - d…

> a lot of my time now is spent reviewing code rather than writing it.

Reviewing has never been a panacea. It’s a best-effort at catching obvious mistakes, like a second opinion. Only with highly rigorous tests can reviewing give as high confidence as I trust another engineer or myself. Generally cadence of code output has never been a bottleneck for me, rather the opposite (if I had more time I’d write you a shorter letter).

Most importantly, writing code that is testable on meaningful boundaries is an extremely difficult and delicate art form, which ime is something you really want to get right if possible. Not saying an AI can or can’t do that, only that it’s the hardest part. An army of automated junior engineers still can’t win over the complexity beast that yolo programming causes. At some point code mutations will cause more problems as side effects than what they fix.

Post reply on HN