MrDarcy

6 ms·

61.
by MrDarcy 9mo ago
The best wolves I know define their own process to find out if the idea is good and ignore all externally imposed processes that get in their way.
62.
by MrDarcy 9mo ago
No I haven’t. I’ve also deliberately avoided the types of places where that happens throughout my entire career. Those types of ideas and projects have no need for the approval of authority. They are their own authority. If a person in au
63.
by MrDarcy 9mo ago
This. And the reason this is relevant today is because an excellent build and test system multiplies the effectiveness of coding agents which multiply again the effectiveness of next level engineers. There aren’t many other things I can thi
64.
by MrDarcy 9mo ago
The wolf and Rands are next level. > Good work does not speak for itself, it needs a shitload of people to speak for it. Good work does speak for itself. The work speaks directly to other like minded engineers who "get it" and
65.
by MrDarcy 9mo ago
On the other hand the incorrect values may drive architects to think more critically about what their tools are producing.
66.
by MrDarcy 9mo ago
https://landscape.cncf.io/ are micro services
67.
by MrDarcy 9mo ago
The ecosystem. Micro services are the most efficient way to integrate CNCF projects deeply with your platform at any size. Edit: Genuinely curious about the downvotes here. The concept directly maps to all the reasons the article author cit
68.
by MrDarcy 9mo ago
I know at least one of the companies behind a coding agent we all have heard of has called in human experts to clean up their vibe coded IAC mess created in the last year.
69.
by MrDarcy 9mo ago
The specific point I was trying to make was along the lines of, "I, a seasoned network security and red-team-type person, could have done this in Wireshark without AI assistance. And yet, I’d probably lose a bet on a race against someo
70.
by MrDarcy 9mo ago
While my reply was snarky I am prepared to take a reasonable bet with a reasonable test case. And pay out. Why I think I’d win the bet is I’m proficient with tcpdump and wireshark and I’m reasonably confident that running to a frontier mod
71.
by MrDarcy 9mo ago
TIL serializing a protobuf is only 5 times slower than copying memory, which is way faster than I thought it’d be. Impressive given all the other nice things protobuf offers to development teams.
72.
by MrDarcy 9mo ago
How much are you staking on that bet?
73.
by MrDarcy 9mo ago
Terrible for those laid off but perhaps not for Evernote customers if it means there isn’t unwelcome feature creep.
74.
by MrDarcy 9mo ago
Google, Amazon, and Microsoft aren’t profitable?
75.
by MrDarcy 9mo ago
Go for it. They have allow and deny lists.
76.
by MrDarcy 9mo ago
Two very useful use cases for rebase: 1) rewrite history to organize the change list clearly. 2) stacked pull requests of arbitrary depth. You’ve never run a bisect to identify which commit introduced a specific behavior? This is when I’ve
77.
by MrDarcy 9mo ago
This comment confuses privacy with anonymity.
78.
by MrDarcy 9mo ago
The best solution I’ve found for probes is to put all eggs into the basket listening on 443. Haproxy with SNI routing was simple and worked well for many years for me. Istio installed on a single node Talos VM currently works very well for
79.
by MrDarcy 9mo ago
I get this and operated this way for most of 2025. In Q4 I paired up with peer who has similar experience level to me. They specialize in TypeScript, I specialize in Go, we're both ops / platform grey beards. This pattern of LLM
80.
by MrDarcy 10mo ago
So much this. The act of having the agent create a research report first, a detailed plan second, then maybe implement it is itself fun and enjoyable. The implementation is the tedious part these days, the pie in the sky research and planni
81.
by MrDarcy 10mo ago
TLS SNI routing has fixed the multiple authorities listening on one IPv4 address port 443. Most ISP’s implement IPv6 by using the single IPv4 address as a v6 prefix. This results in the entire LAN needing to change local addresses every tim
82.
by MrDarcy 10mo ago
For my part I'm obviously at the "I'm different" level and I don't feel insulted discussing the whole spectrum. Hopefully that's true for more people over time. Edit: In retrospect I suppose the “developed wro
83.
by MrDarcy 10mo ago
I agree with you and I’m struggling to see how my reply wasn’t helpful. I’m saying those people who get run over by a car in this analogy shouldn’t be run over by cars on the highway. I look at what’s wrong with the highway. I don’t believ
84.
by MrDarcy 10mo ago
Not that it's about me but my guess was 20%, 1/5, not far off from 1/4. I know and am close to many autistic people, those with and without high support needs. That estimate is of diagnosed people. Many autistic people are
85.
by MrDarcy 10mo ago
This comment sparked curiosity about the distribution of support needs among autistic people. My hypothesis is those people with high support needs are in the minority and those without are the plurality. After digging into it, the hypothes
86.
by MrDarcy 10mo ago
Downvoted for “developed wrong” when it’s becoming increasingly clear Autism isn’t so clearly a disorder or wrong as it is different.
87.
by MrDarcy 10mo ago
What is the purpose of putting the pid file into /dev/shm ? I’ve never seen that before and am curious to learn more about the technique.
88.
by MrDarcy 10mo ago
They are building data centers of TPUs now, not general purpose processors.
89.
by MrDarcy 10mo ago
As a hiring manager, if someone shared a profile of a person with a PHD in earth sciences and 8 years of “full stack” experience I’d immediately assume they were a generalist with surface level familiarity and no mastery or depth in any one
90.
by MrDarcy 10mo ago
Who cares? Nobody is building large scale inference services with macs.

More ›