rco8786

15 ms·

271.
by rco8786 1y ago
I just get a generic error every time I try.
272.
by rco8786 1y ago
This entire post is a lie
273.
by rco8786 1y ago
I still don’t even know what an agent is. Everyone seems to have their own definition. And invariably it’s generic vagaries about architecture, responsibilities of the LLM, sub-agents, comparisons to workflows, etc. But still not once have
274.
by rco8786 1y ago
Does it? RAM is not particularly expensive
275.
by rco8786 1y ago
Monolith architectures are not constrained to running as a single process or on a single machine at all.
276.
by rco8786 1y ago
That was me. Self taught since I was 13ish. Landed a job on a team of "actual" engineers (think Stanford/MIT grads doing massive scale distributed systems in the 2010s) and became painfully aware of all I had missed in a CS d
277.
by rco8786 1y ago
What can you not do in a monolith? You can still have async queues and different event processors that stop and start independently within a monolithic deployment.
278.
by rco8786 1y ago
It's a neat idea, but having a hard time getting useful results (the forever plight of AI). I asked about a road trip from Atlanta to Houston with 3-5 interesting stops along the way. First it sent me to south Georgia, then to northern
279.
by rco8786 1y ago
Yea I'm not trying to take away from any of that, nor provide generalized advice. Only addressing the crowd of people who will shy away from pursuing (or poo poo others) when their idea isn't some sort of original innovation that
280.
by rco8786 1y ago
You are confusing technology for product. The product space they entered, UX design tooling, was very crowded. Sketch and InVision were the kings and there were a dozen other reasonable options to choose from (including ones in browser with
281.
by rco8786 1y ago
That was certainly not my intention. Nothing but respect for what they’ve built. Only pointing out that they entered a crowded space that already had “winners” in it, and succeeded.
282.
by rco8786 1y ago
Honestly the cycle continuing is a breath of fresh air relative to the AI-ification of our entire industry.
283.
by rco8786 1y ago
Congrats to the team here. Let it be a lesson for anyone worried that “their idea has been taken” or “there already solutions for this” out there.
284.
by rco8786 1y ago
You can already do this in .Claude/settings.json
285.
by rco8786 1y ago
One major difference is that MCP has discovery built into the protocol. There’s nothing in REST that informs clients what the API can do, what resources are available, etc.
286.
Ask HN: Has anyone managed to implement OAuth on an MCP server?
1 points by rco8786 1y ago|2 comments
287.
by rco8786 1y ago
I have a pet theory that most "new" dinosaurs being discovered are just dinosaurs we already know about at different stages of life.
288.
by rco8786 1y ago
> I'm not familiar with the texas government > the "narrowly focused, dogmatic" politicians run the two chambers? Yes. This is VERY Texas. The state government is 100% conservative republican. "Texas is a majority
289.
by rco8786 1y ago
Standardization. You spin up a server that conforms to MCP, every LLM instantly knows how to use it.
290.
by rco8786 1y ago
An apartment building a few blocks over from me has these. It makes the whole area look like a night club. Very weird.
291.
by rco8786 1y ago
Genuine question, was this reply generated or in any way augmented by AI?
292.
by rco8786 1y ago
Yep totally agree. And also realistically once we hit that kind of scale it was perfectly reasonably to start looking at other options as well. It's been a while but I think we were at like 250mm MAUs when we finally decided to break u
293.
by rco8786 1y ago
> Then they realized it was hard to scale A rumor perpetuated by exactly one company - Twitter. I know because I was there when it happened and helped dismantle the original monolith. Rails scales just fine for 99%+ of business cases. If
294.
by rco8786 1y ago
This doesn't sound significantly different than any other large tech org. If your data/service/api is used by a lot of other people in the org, you have to work with them to make sure your change doesn't break them. That
295.
by rco8786 1y ago
> If a human therapist told them that, they'd be going to jail. Is that actually true? I know zero about the legal bounds of licensed therapists, so genuine question.
296.
by rco8786 1y ago
Yes. “Basic health checks” is not a real thing. I mean that genuinely. > What makes you think it’s hard? Being responsible (or rather, on a team of people responsible) for a status page of a big tech co made me think it’s hard. “Is it do
297.
by rco8786 1y ago
Google is 10 minutes late updating their status page. "So shady" It's really, really hard to make a status page realtime.
298.
by rco8786 1y ago
> Instead, use modular methods like retrieval-augmented generation, adapters, or prompt-engineering — these techniques inject new information without damaging the underlying model’s carefully built ecosystem. So obviously this is what mo
299.
by rco8786 1y ago
What’s not clear to me is how agents/AI written code solves the “half hearted review” problem. People don’t like to do code reviews because it sucks. It’s tedious and boring. I genuinely hope that we’re not giving up the fun parts of s
300.
by rco8786 1y ago
There's just so much that feels "off" here, it's hard to put my finger on any particular part of it. Fundamentally, we acknowledge that AI writes crappy code and there's no visible path forward to really change that

More ›