"no worries man I know what I'm doing" it’s first sentence in AI-using human mouth that tells me to turn around and run.
I’m using models, I’ve built two whole projects without looking at more than 1% of code, but I’ve tested the interface and went through data on disk to verify the assumptions.
Sad to see you have a FRIEND who behaved this way.
> I’m using models, I’ve built two whole projects without looking at more than 1% of code, but I’ve tested the interface and went through data on disk to verify the assumptions.
so, would you say "no worries man I know what I'm doing" about how you use AI?
what does "verify the data on disk" even mean? I guess without knowing what the project is it's weird to think testing it + checking some files is enough, at least for me
I would love to know what Claude model was used to one-shot the gym app.
My observations lately show that only top model is able to do that, a e.g. Opus got dumber after Fable premiere.
Second that question not only to OP/GP.
Comparing Scylla and Cassandra - the choice is obvious, but current offering is a complete no-go.
Also Mongo is catching up in the use case we have and it’s even more scary. We are all the time at the exact edge of Mongo scalability and I’m talking about very good, few-years-invested-in setup on NMVe that is really delicate…
So you are already sometimes being the „meat proxy”. It only shows one much worse thing - the people asking you are first to blame/for, because they didn’t ask LLM. Maybe it’s temporary and it will change.
One added value I see in senior knowing how things work is simple - in case of outage or business incidents (caused by incorrect logic produced by LLMs from even worse prompts or incomplete spec) senior can act faster and answer difficult questions live. Now we are going to have bunch of kids looking at „Hallucinating…” and „Crashing the prod…” status bar in CC CLI.
Maybe this is what’s gonna be acceptable and it’s just new world.
We observe 200+ (can’t say closer number) purchases per second for single SKU with good marketing and price.
The other thing that bothers me - why not real stable, but maybe „too old, medieval” solution with Redis as the main source of through - without any sync with SQL at all in terms of stock… it worked in my previous job with much higher traffic (1000s/s). Yup, we ended up with app-side sharding, but it was stupid-simple.
You new to the world, right? ClickHouse proven being a beast for analytics, tracing and even some basic metrics storage (VM won of course, but still CH is probably 2nd or 3rd)
It’s still a thing in big corps. We have all business nanoservices (I can’t call them micro) at JVM (moving to Kotlin from Java) and any try using Go is silently killed by those JVM talibans.
I wonder if no one really thought there about rewrite that will still connect with all old dependencies with old protocols, but also will expose new kind of API that can be used by next rewritten block.
We also had mainframe, we used Oracle. It turned out that casual server is also reliable in long term and we restarted them only for upgrades of OS. Of course moving to some distributed, fault tolerant option also feels like good direction, but instead not-doing the perfect rewrite at once you can do it two/three step migration. Costs nowadays? You already fixed that.
While reading all about cameras and other sensor I would try to use distance-driven sensors every where. Standing pin vs laying one have enough height difference to not mislead even simple distance echo-based sensor. Other techs I see more reliable are LIDARs, because they don't require lighting. Pure camera should have enough light to see the pins (or you have to do some IR based markings and IR camera) or you can't make the space darker for events etc.
reply