The memory forking is really interesting. I wonder if copy-on-write at the VM level, O(1) with respect to machine size, won't scale cost with how many forks to take, but 320ms median seems good for the branch-and-explore pattern without reprovisioning every time.
One gap I'm noticing in these comments and in the current sandbox landscape is Windows. Every platform mentioned in these comments like E2B, Daytona, Fly Sprites, Sandflare appears Linux-native. Makes sense for coding agents targeting Debian environments, but a real category exists to automate Windows-specific workflows: enterprise software, ERP systems, anything that runs only on Windows.
If anyone wants to run agents in Mac or Linux and need to access Windows for computer use, Dexbox could be helpful. [github.com/getdexbox/dexbox]
I launched an open source developer tool called Dexbox to run agent workloads that quickly provision and run Windows desktops. It's a CLI and MCP experience that's different from Freestyle, but slightly closer to our Windows-specific production infra, Nen. I like Freestyle's cool UI that shows off the unique technical approach and developer friendliness. Nen's a bit closer to that experience.
Isn’t frontend more complex? If my task starts with a Figma UI design, how well does a code agent do at generating working code that looks right, and iterate on it (presuming some browser MCP)? Some automated tests seem enough for an genetic loop on backend.
That seems unusual. I would assume trivial means the default approach works for most cases. Perhaps mutable global variables are not a common use case. Unsafe might make it easier, but it’s not obvious and probably undesired.
I don’t know Rust, but I’ve heard pockets of unsafe code in a code base can make it hard to trust in Rust’s guarantees. The compromise feels like the language didn’t actually solve anything.
I have hardly come across aspiring-CEOs who do not have much ego and listen to other founders, let alone founders or serial-founders. After some founder dating last year, I am a bit jaded, though.
Morality is not about intelligence, but some folks believe living by morals is submitting oneself to the lack of intelligence (or stupidity even). The real [evil] geniuses prune their decision tree by not considering morality so they maintain optimum efficiency.
There is a lot of sentiment in these comments about needing to scale still. I wonder how many need to do this in a pre-PMF stage vs growth stage? The trade off is faster growth if your PMF bet wins and loss of time if your bet goes south.
Maybe not capabilities, but I'm wondering if Postgres has gotten any easier to scale horizontally. The administrative overhead of scaling and maintenance with MongoDB seemed lower to Postgres to me.
Would love to hear from others with more Postgres than I.
Very wealthy people have access to very exclusive and potentially unethical and/or illegal capabilities. Many do not think they are bound to the same rules as the rest of society. Some are probably not.
OpenAI the company may not be responsible, but there are elites in the OpenAI sphere who might be.
> why structured editors haven’t taken over the world: most programmers find them so annoying to use in some situations that they don’t view the pros as outweighing the cons.
Seems like "annoying" refers to a user interface annoyance.
I'm guessing the following since I couldn't tell what structured editing is like from the article:
Keyboard entry is immediate, but prone to breaking the program structure. Structured editing through specific commands is an abstraction on top of key entry (or mouse), both of which add a layer of resistance. Another layer might come from having to recall the commands, or if recognizing them, having to peruse a list of them, at least while learning it.
What does the developer's experience with incremental parsing feel like?
This article was kind of lazy. It’s a lot of conjecture, a couple anecdotal observations, and some rhetoric for good measure. Basically, the content was a let down from the title. On the other hand, I didn’t know Data Scientist candidates would also be tested with LeetCode questions.
The opposite seems true (regarding Go being over-fitted to Google problems).
It was designed for Google, but Google has not adopted it widely. Google is still heavily C++ and Java after all these years. The outside world loves it way more. Kubernetes isn't used internally at Google (except Cloud, which is not any different from any other cloud provider) though it's sponsored by Google. Bazel is probably in a similar boat.
Being designed to solve Google problems doesn't mean that it actually solves Google problems well, or that Google thinks it does.
I became introduced to Go through my startup and my experience was it a delight to work with in very small teams.