yandie

5 ms·

31.
by yandie 1y ago
It's actually an important practice to have a docker image cache in the middle. You never know if an upstream image is purged randomly from docker, and your K8s node gets replaced, and now can't pull the base image for your servic
32.
Mapped: The Compute, Cash, and Contracts That Power OpenAI(visualcapitalist.com)
2 points by yandie 1y ago|0 comments
33.
by yandie 1y ago
This isn't just context optimization. Not much different from agent-to-agent workflow IMO.
34.
by yandie 1y ago
> I'm pretty sure a majority of us aren't really on a "side", Many of us don't vote either. And our two party systems have created extreme partisanship. I wish it could be different because I do love this country
35.
by yandie 1y ago
Why do you need to be a billionaire? That's a very different goal from retiring comfortably by the way.
36.
by yandie 1y ago
And it’s still slow on my phone (and I’m not a logged in user)
37.
by yandie 1y ago
This Netflix blog isn’t yahoo scale either
38.
by yandie 1y ago
I can’t imagine the number of editors is that large? Are we talking about hundreds or thousands? Taking a look at the site there can’t be that many users that need this low latency preview capability.
39.
by yandie 1y ago
But I need to build a beautiful system with global scale!! They inflated the problem of “make content preview faster” for a small number of users to “make a fast global cache system”. That’s promotion material for you
40.
by yandie 1y ago
They probably saved the equivalent of an engineer's salary!!
41.
by yandie 1y ago
I wonder how this would apply with vision models? I tried with a few example of single images and they appear to do well. I did a few toy examples and they seem to do pretty well (Claude + Gemini) with spotting differences. An example image
42.
by yandie 1y ago
We were working with folks from other central ML infra teams in other companies. Amazon is the one that doesn't have any funding for central ML infra (they have the central infra for software). Having interacted with many companies aft
43.
by yandie 1y ago
I was with Amazon but wasn't part of Alexa. I was working closely with the Alexa team however. I remember vividly the challenge of building centralized infra for ML at Amazon: we had to align with our organization's "success
44.
by yandie 1y ago
Per my experience with GenAI legal teams, that’s a no go. It’s not been tested in court though
45.
by yandie 1y ago
I don't think this decision can force the Department of State to issue new visas for Havard students unfortunately. At least existing students *might* be alright...
46.
by yandie 2y ago
I've been a long term Alexa user and kinda hard to get out of the ecosystem, but it's gone downhill fast in the last few years in terms of the UX. * It forgets about my configured routine ("good night" -> "turn o
47.
by yandie 2y ago
European country really gave me bad taste when it comes to visa experience. Gave me the absolute minimum despite having a good paying job - and even after marrying a EU citizen I still had a hard time. This is news because this woman happen
48.
by yandie 2y ago
And you need to get work authorization (EAD card) for that. It’s not a given
49.
by yandie 2y ago
When I was at a big tech, the interpretation by their lawyers was that by running AGPL, we will have to open source everything in the network to users. The problem is in the definition of "Modification" - Per [AGPL]( https:/&
50.
by yandie 2y ago
You can come on L1B and then try to get into the H1B lottery (will be tough since you don't have a US degree - I assume). I decided to go with L1B since H1B was impossible to get back then and the timeline worked in my favor. I came to
51.
by yandie 2y ago
Hey congrats! Looks like a really cool idea. Looks like you're pushing for the throughput angle - that could be important but IMO it's not often you come across devs who need this level of throughput without dealing with large sca
52.
by yandie 2y ago
I've a very mixed accent and it just pretty much throw a random country in SE Asia (correct region) with different tries.
53.
by yandie 2y ago
Having my app with direct write access to the scheduling backend could be... problematic from the separation of concerns. We'll have to deal with connection pooling etc and we typically prefer to separate DB logic (interaction with pos
54.
by yandie 2y ago
Just tried the deep search. It's a much much slower experience than perplexity at the moment - taking sweet many minutes to return result. Maybe it's more extensive but I use perplexity for quick information summary a lot and this
55.
by yandie 2y ago
Temporal can use postgres as a backend. What makes this product different?
56.
by yandie 2y ago
The inverse of OceanGate? Call it StratosphereGate or AirGate?
57.
by yandie 2y ago
This explains the Multi acquisition.
58.
by yandie 2y ago
You need to hover "Why am I seeing this destination?" - the LLM is saying it's returning Delta Airlines - not obvious. The RAG flow obviously can't pull Delta Airlines, but the explanations all mention Delta (depending o
59.
Alaska Launches a Chatbot That Hallucinates about Delta Airlines(alaskaair.com)
6 points by yandie 2y ago|3 comments
60.
Observability is yesterday – GenAI needs a control center(businesswire.com)
1 points by yandie 2y ago|0 comments

More ›