So based on the comments, I'm the only person who think our jobs will be diving into years-old LLM-generated code to fix it all, and laughing: "Oh my, do you all remember when the entire industry burned 5 years and b/trillions of dollars just to try to force a use case to work that wasn't even the right match for the tools?"
I'm not even trying to be snarky. I just see that LLMs already work fine in almost every other use case. Better than fine - they have been implemented and they work, and don't need much more advancement. But coders are still fighting it, coming up with more and more elaborate setups, spending more and more money, just to get to a level of coding that is passable. Not great, not better than human, just a "CRUD apps work." level of passable.
Once the exec teams stop pushing the FOMO from down on top, I think people will settle down into a much lower level of AI assistance. It will continue to be used, but we'll have it boilerplate the simple stuff out for us, while we focus on the meatier problems. And that is not all that different than pre-AI: whip the easy stuff out, then work on the hard stuff.
Things that I reckon will become a lot more important from a developer's perspective over the next year:
- Shaping work so it is more decomposable, legible, verifiable and understandable
- Property testing, formal verification (exhaustive proofs, and bounded model checking), and test verification (e.g. mutation testing)
- Understanding product market fit and opportunity space
- Application of product management techniques (e.g., constraint, risk, and management by exception)
- Hypothesis forming, validation and falsification
- Evaluation of systems (agentic systems and toolsets used by agents)
- Effectively shaping domain knowledge into agent context in a way that stays current
- A greater awareness of o11y and cost recovery
- User experience, developer experience and agent experience
I'm not suggesting that developers should replace UX engineers or product managers (in fact, I think that demand for experts in these fields will grow with the volume of software). I am also not suggesting that these growth areas are not things developers need to be mindful of now, but that the need for by default application of automated development rigor will grow.
I see this as more or less analogous to the DevOps shift of the mid 2010s. Developers will grow their horizons, become comfortable with achieving more and moving faster safely, and ultimately put more trust in observable and legible delegated systems rather than doing things by hand.
It wont be implementing current tech using AI for sure. Everyone and their mothers will be using AI at OS or browser level. For example, What if users block ads using AI? Companies will have figure out a way to stay ahead and continue to serve ads.
To be relevant in the software industry, we need to get upgraded all the time similar to an app. According to me, with the more development in AI side, it will be more like managerial job instead of proper dev job. You will be managing your agents, assigning them tasks so your technical knowledge should be up to the par so you can review them properly and keep yourself relevant. The development will be quick but still needs to be manually tested and manually approved like it is right now because the models will never be deterministic so they will require reviewers and managers.
I agree but with the recent developments, what if the agents are able to manage as well like the orchestration agents then how would the scenario change?
Even if the LLMs are able to do managerial work, still they will need supervisors on top of them. According to me, it will more like team leads and managers with LLMs as their developers or maybe planners.
Another thing, it also depends on LLM costs as a lot of companies are exceeding their budgets and with more dependency on LLMs, the costs will be even more which can eventually reduce the company profits.
Developers at Mistral AI in France are already no longer writing lines of code. The future of developer work lies in understanding the client's business in order to create a custom application tailored to their needs.
To me, that seems the rough description of a product manager!
Also something that requires heavy social skills. What will we, the vast nerd majority do?
I mean many developers are already out of job and most will follow. I mean we are going from being skilled workers to unskilled workers just in a few years.
I think the majority of dev work will be gone. We already demonstrated just throwing more compute at it is leading to better and better results. We'll likely have advancements that make it so regular people that ask the right questions and are diligent can do end-to-end app creation and management. I think it's also likely that standardizations for security and account management and compliance will just become skills or other types of automations that the drastically more competent agents of 2027-2028 can simply utilize.
It is seeming less and less unlikely for this to happen.
So many businesses right now online you can simply have Claude Fable copy and implement in an hour. That'll probably scale up as well in the future. I just cloned Piano Marvel earlier today in an hour and added missing features that I really wanted on the site that aren't there.
I think society will be a lot crappier for people that don't already have assets and investments that outpace inflation.
So based on the comments, I'm the only person who think our jobs will be diving into years-old LLM-generated code to fix it all, and laughing: "Oh my, do you all remember when the entire industry burned 5 years and b/trillions of dollars just to try to force a use case to work that wasn't even the right match for the tools?"
I'm not even trying to be snarky. I just see that LLMs already work fine in almost every other use case. Better than fine - they have been implemented and they work, and don't need much more advancement. But coders are still fighting it, coming up with more and more elaborate setups, spending more and more money, just to get to a level of coding that is passable. Not great, not better than human, just a "CRUD apps work." level of passable.
Once the exec teams stop pushing the FOMO from down on top, I think people will settle down into a much lower level of AI assistance. It will continue to be used, but we'll have it boilerplate the simple stuff out for us, while we focus on the meatier problems. And that is not all that different than pre-AI: whip the easy stuff out, then work on the hard stuff.
Things that I reckon will become a lot more important from a developer's perspective over the next year:
- Shaping work so it is more decomposable, legible, verifiable and understandable
- Property testing, formal verification (exhaustive proofs, and bounded model checking), and test verification (e.g. mutation testing)
- Understanding product market fit and opportunity space
- Application of product management techniques (e.g., constraint, risk, and management by exception)
- Hypothesis forming, validation and falsification
- Evaluation of systems (agentic systems and toolsets used by agents)
- Effectively shaping domain knowledge into agent context in a way that stays current
- A greater awareness of o11y and cost recovery
- User experience, developer experience and agent experience
I'm not suggesting that developers should replace UX engineers or product managers (in fact, I think that demand for experts in these fields will grow with the volume of software). I am also not suggesting that these growth areas are not things developers need to be mindful of now, but that the need for by default application of automated development rigor will grow.
I see this as more or less analogous to the DevOps shift of the mid 2010s. Developers will grow their horizons, become comfortable with achieving more and moving faster safely, and ultimately put more trust in observable and legible delegated systems rather than doing things by hand.
It wont be implementing current tech using AI for sure. Everyone and their mothers will be using AI at OS or browser level. For example, What if users block ads using AI? Companies will have figure out a way to stay ahead and continue to serve ads.
this is a rosy view; the AI will have the ads, just like good search had the ads.
The only way you're opting out of ads is local AI.
To be relevant in the software industry, we need to get upgraded all the time similar to an app. According to me, with the more development in AI side, it will be more like managerial job instead of proper dev job. You will be managing your agents, assigning them tasks so your technical knowledge should be up to the par so you can review them properly and keep yourself relevant. The development will be quick but still needs to be manually tested and manually approved like it is right now because the models will never be deterministic so they will require reviewers and managers.
I agree but with the recent developments, what if the agents are able to manage as well like the orchestration agents then how would the scenario change?
Even if the LLMs are able to do managerial work, still they will need supervisors on top of them. According to me, it will more like team leads and managers with LLMs as their developers or maybe planners. Another thing, it also depends on LLM costs as a lot of companies are exceeding their budgets and with more dependency on LLMs, the costs will be even more which can eventually reduce the company profits.
Totally agreed
Developers at Mistral AI in France are already no longer writing lines of code. The future of developer work lies in understanding the client's business in order to create a custom application tailored to their needs.
To me, that seems the rough description of a product manager! Also something that requires heavy social skills. What will we, the vast nerd majority do?
That is what my job as a developer has always been.
I mean many developers are already out of job and most will follow. I mean we are going from being skilled workers to unskilled workers just in a few years.
I think the majority of dev work will be gone. We already demonstrated just throwing more compute at it is leading to better and better results. We'll likely have advancements that make it so regular people that ask the right questions and are diligent can do end-to-end app creation and management. I think it's also likely that standardizations for security and account management and compliance will just become skills or other types of automations that the drastically more competent agents of 2027-2028 can simply utilize.
It is seeming less and less unlikely for this to happen.
So many businesses right now online you can simply have Claude Fable copy and implement in an hour. That'll probably scale up as well in the future. I just cloned Piano Marvel earlier today in an hour and added missing features that I really wanted on the site that aren't there.
I think society will be a lot crappier for people that don't already have assets and investments that outpace inflation.