From my experience, there are mainly 3 burnout reasons.
1. Multi-tasking is the top one. I usually have to frequently switch between 3 to 5 agent windows which are on different things. It's extremely exhausting when each round takes a few minutes. Before coding agent era, I believe most developers had chance to spend 2+ hours focusing on one thing. Now coding agents have increased my spectrum on the tech stack, but the bandwidth to do deep work isn't increased.
2. Agents are good at getting things running without crash, but do not guarantee to produce correct code. This is quite different from human experts with fundamental knowledge.
3. I also get frustrated when reviewing piles of AI generated low quality PRs. My attention is a limited resource. I don't waste too much energy on other people's work, but if I don't spend more effort, the entire project is corrupted quickly by reckless AI generated code without human author's careful thoughts and designs. Working with people who have less due diligence in mind is painful, working with them in coding agent era is 10x painful because they produce 10x shit. It's a team culture challenge that cannot be easily enforced.
Agreed. I am working hard to restrict myself to only 1-2 agent workflows at a time. More is untenable, though it’s so easy to fall into the trap of deploying an agent “just for this minor fix.”
The worst is that every agent session is generating so many "btw fix this" side-quests that it is really hard to stay in the task focus. I throw some into todo list manually but still it is exploding by the day. Perfect is enemy of good.
apples private cloud compute can get close, its still not 100 safe because backdoors and crypto breaks are possible but you go from trusting the data center operator with all their employees to only the person thats inspecting new hardware and giving out certificates (apple in this case). if some well known non profit like mozilla or isrg starts doing it with full open source software its like the best possible security
The handy thing about trusting Apple here is: you were already trusting Apple. I don't necessarily like that trust works this way, but that's just the physics of it.
Just like most things in life the guarantee it based on the entity/person providing said guarantee.
I can host a LLM in my basement and guarantee it, but would you trust me? Now you can say that you don't trust any company, but B2B relies on counterparty risk.
That is completely obvious, it’s like saying “100% security does not exist”.
I believe you are falling into the nirvana fallacy: No shades of grey, if it’s not perfect it’s as bad as the rest.
This is a very inefficient way of thinking as it is not possible to self host everything for most people, it just demands too much time.
Hence its is a perfectly valid approach in my opinion to looks at better (or, very often, “less worse”) SaaS solution.
If they states ZDR on a model, the likeliness of it leaking less data to some LLM data training is higher simply. If the business model of a company is built around a differentiator which is data privacy, that also significantly increases probability that data is not being leaked/sold.
It’s all grey, relative and about probabilities. Nothing’s perfect – another captain obvious thing.
Same here. Don't understand why Github hasn't supported this until now. I'm tired of reviewing PRs with thousands of lines of changes, which are getting worse nowadays with vibe coding.
Right. Yesterday I tried a simple task that just adds Required[] notation to all class fields. After making the change on one field, Cursor allows me to press tabs and update all other fields. VSCode doesn't understand what I was trying to do after the first operation, which is surprisingly bad (no improvement after months). Also I'm not in favor of the conversational experience of claude code or other CLIs for such trivial task.
I'd be happy to know what else can provide a better user experience than Cursor.
Disclaimer: I get enterprise level subscriptions to these services via my employer. I personally don't pay for them and never consider their cost, if that matters.
I tried to set up Dart runtime and run cli on a Linux server. Then I realized that it just prints a message and quits. Maybe it's better to directly tell user cli isn't implemented yet.
Not only that, by the time they figure how to do it the rest of the world will have moved on by at least 2 generations. SMIC has announced 14nm (but shipments in volume aren't clear), TSMC will have 5nm chips shipping in volume by the end of the month.
how does that matter? SMIC had the 7nm news recently, not sure about that but even if it is 14nm, can't they throw money on the problem and get something to start with?
they don't have to achieve A100 levels in the first go, even Rpi level chips should be enough to give them experience
I guess they are not able to buy the tooling (think ASML EUV) to make advanced node either. So, they’d also need to throw a shitload of time and money at this.
This is one of those problems that “throwing money at it” doesn’t speed it up. China has been throwing money at the problem for over a decade and they’re still behind.
reply