TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Ask HN: Can Devin genuinely replace the roles of developers?

29 pointsby sunnysograabout 1 year ago
These days, everybody is talking about Devin and its threats to developers. I would like to gather feedback from others to gain a comprehensive perspective on this matter. Do share your thoughts.

34 comments

fhd2about 1 year ago
My perspective (as a CTO who&#x27;s hired hundreds of developers in the past 12 years) is that I don&#x27;t have a use case for Devin, from what I&#x27;ve seen about it.<p>That comes down to why I hire developers in the first place: To share my responsibilities with people I can trust.<p>I don&#x27;t hire them to write code or to close tickets. The act of programming, I consider an exercise that helps them understand the problems we solve and the logic of our solutions. I&#x27;m always excited when I have a well specified ticket I can hand to a new hire to learn the ropes. So the kind of thing I can imagine Devin can pull off at some point, that&#x27;d actually be detrimental to the kinds of teams I build.<p>I don&#x27;t think I represent the majority of why people hire developers though, so I guess tools like that may well have a big impact on the industry. Nobody can predict that though.<p>Uncertainty sucks, but it&#x27;s how things are. I find the best way to deal with uncertainty is to become better at adapting to unforeseen circumstances. Programmers have quite a bit of experience with that, for what it&#x27;s worth.
评论 #40209029 未加载
sk11001about 1 year ago
&gt; These days, everybody is talking about Devin and its threats to developers.<p>If by these days you mean the two days after it came out several weeks ago, and by everyone you mean a handful of people on social media, sure.
评论 #40241212 未加载
hathymabout 1 year ago
Devin was debunked as a fraud: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tNmgmwEtoWE&amp;t=1363s" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tNmgmwEtoWE&amp;t=1363s</a>
评论 #40209119 未加载
评论 #40208880 未加载
评论 #40211292 未加载
评论 #40208864 未加载
lewdevabout 1 year ago
The way I see AI programming assistants is that it would help juniors be a bit more productive, but senior developers can do with out the assistant.<p>I&#x27;ve used Cody and Copilot and it just gets in the way because I know exactly what I need to write and neither really helped me.
评论 #40211821 未加载
评论 #40208664 未加载
评论 #40208679 未加载
评论 #40208863 未加载
TowerTallabout 1 year ago
You should watch this video that shows that some of the Devin demos are fake. <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tNmgmwEtoWE&amp;feature=youtu.be" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tNmgmwEtoWE&amp;feature=youtu.be</a>
评论 #40209373 未加载
davidgerardabout 1 year ago
Devin? No, it was a rigged demo verging on fraud. See previous discussion: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=40008109">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=40008109</a><p>Future hypothetical AI coding assistants that don&#x27;t exist yet? While I won&#x27;t say it&#x27;s <i>philosophically impossible</i> that they&#x27;ll move beyond extreme autocomplete with security holes, I&#x27;ll say it&#x27;s not up to me to disprove someone else&#x27;s hypothetical. Show me the thing.
yNeolhabout 1 year ago
My take on LLMs is that it won&#x27;t scale much of what we already see because this is &quot;just&quot; text prediction on steroids. I&#x27;m not an expert by any stretch of the imagination, but that&#x27;s my opinion, and going through that path, no, I don&#x27;t see this path as the best path for &quot;autonomous development machines&quot;, only powerful autocompletion like we already see today.
评论 #40208795 未加载
评论 #40208851 未加载
fragmedeabout 1 year ago
The open question is if the answer to your question is &quot;no&quot;, or &quot;not yet&quot;. a bot that takes 1 point jira tickets and goes off and does them would be something, but that currently doesn&#x27;t exist. until it does, we can say it won&#x27;t exist, that it can&#x27;t exist using the current approach, and thus that it won&#x27;t exist within my lifetime. Until it does.
xeornetabout 1 year ago
Depending on what you’re building, likely sometime in the near future. I’m a senior dev, I’m just using Cursor.sh and my day is essentially telling it what I need done - and the code is written much faster than I could have.<p>Essentially I now just architect and review. Cursor has good context, so if that gets extended to the way Devin operates I think this could go pretty far.
评论 #40208931 未加载
评论 #40208765 未加载
padraigfabout 1 year ago
A few thoughts (for context, senior developer, and use chat-gpt every day as an assistant)...<p>In the short-term (5-10 years, I cant see them autonomously producing products), it will need an experienced programmer to interpret and use the output effectively.<p>An implication of this is, in the short-term, developers become even more valuable. You still need them, and these tools will make the developer significantly more productive.<p>I was reading Melanie Mitchell&#x27;s book &#x27;Artificial Intelligence: A guide for thinking humans&#x27; recently (which I&#x27;d recommend). She has this chapter on computer-vision. And as an example, she shows a photograph of a guy in military clothing, wearing a backpack, in what looks like an airport, and he&#x27;s embracing a dog. She makes an insightful point, that our interpretation of this photograph relies a lot on living-in-the-world experience (soldier returning from service, being met by his family dog). And the only way for AI to come close to our interpretation of this, is maybe to have it live in the world, which is obviously not such an easy thing to achieve. Maybe there&#x27;s an analogy there with software development, to develop software for people, there&#x27;s a lot of real-world interaction and understanding required.<p>In terms of autonomously producing products, I see these tools as they are now a bit like software wizards, or a website that Wordpress will create for you. You get a &#x27;product&#x27; up-and-running very quickly, and it looks initially fantastic. But when you want to refine details of it, this is where you get into trouble. AI has an advantage over old-fashioned wizards, in that you can interact with it after the initial run, and refine it that way. But I&#x27;m not sure this is so easy, to have that fine-grained control you have with code. This is where I see the challenge being, to develop tools to talk to it, and refine the product sufficiently.
imafishabout 1 year ago
No, the job will change.<p>It is a tool for building software. You still need to know software development to use the tool.<p>You might not need to actually write code in the future - just like very few write Assembly today.<p>But you still need to know and understand system requirements, systems architectures, integrations, distribution, deployment, maintenance, etc.<p>Software Engineering is more than just coding.
评论 #40208969 未加载
jjiceabout 1 year ago
&gt; These days, everybody is talking about Devin and its threats to developers<p>I think we&#x27;re in different spaces, because I barely hear anything about it. That said, I think the LLM replacing jobs train was blown out of proportion. I heard so much about it replacing developers, but I&#x27;ve seen time and time again it output code with subtle bugs (I&#x27;d argue worse than obvious bugs) and no be able to operate with more than just a little bit of context.<p>I think we&#x27;re in a Pareto distribution situation right now. The majority of getting an LLM to write code was pretty quick to do. To get it to do anything a moderate dev can do will take decades.<p>I&#x27;ve seen it multiple times over the last 15ish months where I&#x27;m reviewing code and I spot a subtle bug in an htaccess file or a bash script that doesn&#x27;t make any sense. The PR comment follow up is then &quot;oh I got it from ChatGPT&quot;. I think these tools become assistants to a human developer who can guide them. That use case is already available and seems to be pretty decent for a lot of folks. Full replacement is so far away that I don&#x27;t have a single thought about it.
bakugoabout 1 year ago
It can replace a junior &quot;developer&quot; who only really knows how to copy-paste from stackoverflow. Anything beyond that, no.
yawboakyeabout 1 year ago
more generally, will developers be replaced? the answer if an affirmative yes. when? unclear at the moment. i think a big moment will be when we free ai&#x2F;automated agents from the shackles of python&#x2F;ruby&#x2F;etc—ie programming languages invented and perfected for the human programmer—and allow them free reins to accomplish their tasks.<p>i contend that given that the universe of computing (ie the capabilities of the processor) is finite, all software is engaged in essentially the same activities (write to memory here, a file over there, etc) such that the difference between any two could be reduced to a matter of interpretation. if so, any automated agent capable of assigning some meaning to these activities should be able to produce a sound program, in whatever computer language, even one very proprietary to the agent.
评论 #40221002 未加载
globular-toastabout 1 year ago
Devin is a replacement for a keyboard, not a developer.
sujayk_33about 1 year ago
The thing that might replace won&#x27;t be Devin, if we are going to make a one-for-all solution, it just won&#x27;t work.<p>If we are talking about a specific use case black box to automate the work, then it might be possible up to a certain limit.<p>because when we are talking about training Neural networks, we are looking for the best numbers. The so-called &quot;emergent abilities&quot; which say that increasing model size makes it smart can be true but what&#x27;s the probability of getting most of the parameters to their correct values? There are billions of them.<p>(Total)Replacement? I highly doubt that.
fergieabout 1 year ago
AI can give suggestions, but it can&#x27;t actually deploy working software into the real world. For the foreseeable future there is always going to be a human who says &quot;yes, lets spin up that elasticsearch cluster that costs 50k a month&quot;.
评论 #40208759 未加载
ips1512about 1 year ago
Recently, it was discovered that Devin was trained on only certain aspects of a problem, enabling it to predict the output accurately. It appears to be specifically trained to resolve similar issues.<p>Although AI is advancing rapidly, if a person learns to adapt in such situations by broadening their learning scope, one can smoothly navigate through such hypes.
Tenokeabout 1 year ago
Junior developers at minimum. Team Leads, no. LLMs are already really good at code, they&#x27;ll obviously get better and better.
margorczynskiabout 1 year ago
It doesn&#x27;t need to replace a developer completely. If it does make development 10-20% faster that&#x27;s on average 10-20% less developers needed to be hired.<p>The current LLMs are not perfect but I recommend anyone to try it out - it really speeds up development, great for creating boilerplate or when trying to use a language you have little knowledge in.
haebomabout 1 year ago
There&#x27;s a certain amount of hyperbole and bluster, but I believe it will happen sooner or later.
screyeabout 1 year ago
Not Devin, it is vaporware, but easy coding will get automated away. If your value as a developer is &#x27;LoC&#x2F;sec&#x27;, then you will be automated away.<p>Architecting, writing requirements, debugging nasty issues and optimizing tricky problems will remain valuable.
评论 #40209009 未加载
kkosteabout 1 year ago
No.<p>Like all the other AI codebots it&#x27;s a tool that can potentially optimize a developers workflow. In the same way that a nail gun optimizes a carpenter&#x27;s workflow. But sometimes the carpenter might just use a hammer.
octocopabout 1 year ago
It&#x27;s going to change how interviews are done with developers, no more leet code homework. And it will speed up some devtime, but I don&#x27;t think its going to replace junior&#x2F;senior devs any time soon.
interbasedabout 1 year ago
I continue to see AI as a tool for developers rather than a replacement. It enables developers to do their job better, but wouldn’t cut it as a standalone developer.
adityaathalyeabout 1 year ago
Like the halting problem, I think LLM tech is a &quot;Halting AI&quot; situation. I wrote about it here last month: <a href="https:&#x2F;&#x2F;www.evalapply.org&#x2F;posts&#x2F;halting-ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.evalapply.org&#x2F;posts&#x2F;halting-ai&#x2F;</a><p>&gt; <i>This riff derives from a recent &quot;AI Programmer&quot; story that&#x27;s making people in my corner of the nerdiverse sit up and talk, at a time when hot new AI happenings have become mundane.</i><p>&gt; ...<p>&gt; <i>It is yet another prompt for me to take the lowkey counterfactual bet against the AI wave, in favour of good old flesh and blood humans, and our chaotic, messy systems.</i>
评论 #40208792 未加载
mellosoulsabout 1 year ago
Given that Devin is hype-ware until it is widely available for testing (remember, its been months since announced and still no sign of a reasonable beta roll-out), it would be better to frame this in terms of the general class that Devin represents, ie. AI-powered &quot;full self driving&quot; coders as opposed to the recent (phenomenal) prompt-driven tools like ChatGPT.<p>So - do FSD coding assistants pose a threat to developers? Sure, just like any tool using GPT3+ class engines do, we are in revolutionary times.<p>But the revolution here is the engines now available thanks to OpenAI and successors, not the wrappers like Devin and others.<p>If you&#x27;re not concerned already, if you&#x27;re not pondering the future already you&#x27;ve missed the point by seeing it only in the likes of Devin.<p>Personally, I think there is both risk and opportunity - it really depends on what sort of mindset you have as to whether you need to feel threatened.
throwaway38375about 1 year ago
No
adezxcabout 1 year ago
I mean, they only made a poor demo since the announcement and then vanished not providing any more info. I do think projects LIKE Devin can replace some developers that work in most popular domains (I&#x27;d guess JavaScript and such), but it won&#x27;t replace devs in complex fields (e.g embedded programmers)
评论 #40208768 未加载
LaurensBERabout 1 year ago
In it&#x27;s current form it&#x27;s rather limited but I can imagine how future version could work as a very productive junior.<p>The real question is, what will be the rate of progress from this point forward?
VoodooJuJuabout 1 year ago
Buy an ad.
throwaway69123about 1 year ago
the u t
user90131313about 1 year ago
no
danenaniaabout 1 year ago
I&#x27;m building a terminal-based tool that is somewhat similar to Devin[1]. Part of the premise of my tool is that &quot;AI software engineer&quot; is the wrong target for current model capabilities.<p>That&#x27;s L5, and it may well be the future, but as it stands today, I believe L3-L4 is a more productive target for a coding agent. Before building full autonomy, we first need the infrastructure to precisely guide the models and iterate efficiently on our interactions with them.<p>Once that foundation is in place, it will then be possible build more and more robust layers of autonomy on top. But crucially, the developer will always be able to a drop a few layers down and take over the controls.<p>1 - <a href="https:&#x2F;&#x2F;github.com&#x2F;plandex-ai&#x2F;plandex">https:&#x2F;&#x2F;github.com&#x2F;plandex-ai&#x2F;plandex</a>