sandGorgon

11 ms·

91.
by sandGorgon 3y ago
This is google+cloudinary published benchmarks of jpeg-xl vs avif vs webp https://storage.googleapis.com/avif-comparison/index.html https://cloudinary.com/blog/contemplating-codec-comparisons
92.
Apache Arrow Flight SQL Adapter for PostgreSQL - Arrow wire format for SQL DB(postgresql.org)
5 points by sandGorgon 3y ago|0 comments
93.
by sandGorgon 3y ago
how is Google doing this ? i mean as opposed to Apple, ChromeOS has to support a much wider variety of chipsets/motherboards/wifi drivers, etc. how is Google guaranteeing 10 years of updates on Linux while supporting newer and new
94.
Introducing Infrastructure Manager: Google Cloud Resources with Terraform(cloud.google.com)
3 points by sandGorgon 3y ago|1 comments
95.
by sandGorgon 3y ago
Harry Potter's golden egg of the merpeople!
96.
by sandGorgon 3y ago
would love your thoughts on this as well - https://github.com/arakoodev/edgechains got frustrated in the same way with "Black Box Prompting - every library hides prompts/chains in layers of libraries...while
97.
by sandGorgon 3y ago
this is very interesting. you had better results here than the openai ada02 and other embeddings like bge ?
98.
by sandGorgon 3y ago
genuinely curious - how does Typescript help in autocomplete ? i thought it was equally fast using function name.
99.
by sandGorgon 3y ago
well my intention was not to trivialize the effort. I think i was trying to ask an architecture question: if you were to build something like Cuber...would you build it over Argocd ? or build it entirely from the ground up. It would still t
100.
by sandGorgon 3y ago
did they build their own litho machines ? Breaking ASML monopoly here ?
101.
by sandGorgon 3y ago
> All that being said, Cuber definitely feels like a different sort of market fit. but it could have been built on top of Argo right ? I totally get an opinionated out-of-the-box workflow and manifest file. But i would rather see that o
102.
by sandGorgon 3y ago
that's a fair point, but Argocd + Packer is just that right ? > some tool that is easily configurable like this that spawns ephemeral environments and deployments based off a checked in production grade configuration template like y
103.
by sandGorgon 3y ago
genuine question - why this over ArgoCD ? which is a huge ecosystem and all the declarative setups. Plus has clean github integration for incremental deploys, etc
104.
by sandGorgon 3y ago
there is a middle ground now. There are multiple agency-as-a-service companies (like lifeandhalf.com, etc) which basically give you an agency infrastructure without running it yourself. where infrastructure means - account managers, talent
105.
by sandGorgon 3y ago
would htmx have fixed this problem ?
106.
by sandGorgon 3y ago
that is super interesting. Do you find that schema-linking using finetuning works better ?
107.
by sandGorgon 3y ago
do you not use a vector db and embeddings search to get the table structure, etc ?
108.
by sandGorgon 3y ago
this is very interesting. just curious - is the context store specific to the table structure of the target database ? or is it a very generic one. im asking this cos the DIN-SQL algorithm seems to be a chain-of-thought kind of algorithm (w
109.
by sandGorgon 3y ago
well im Indian - so i speak 6 languages :D I've constantly worried about how to get the kid to learn the non-mainstream language. Now i get it!
110.
by sandGorgon 3y ago
agreed. There is an equivalent in Java - RxJava used extensively and android and somewhat on server. Everytime i use it, i get a lot of hate. We use it in EdgeChains https://github.com/arakoodev/edgechains
111.
by sandGorgon 3y ago
this WAS super useful! thank you so much. I would love to follow a running list of your recommendations for different age groups (if u maintain one). One question - when u say "allow watching in foreign languages".do u mean watch
112.
by sandGorgon 3y ago
> I can easily design everyday learning activities based on the handful of shows I’ve (since) vetted what have you vetted? am a new parent, so trying to learn.
113.
by sandGorgon 3y ago
what apps have worked well for you in this way ? Im a new parent and want to get thoughts on what helped the child's development and what didnt. for e.g. do you play sesame street, etc type content on the ipad ?
114.
by sandGorgon 3y ago
nice! is this is cloudflare worker & block storage built in Java ?
115.
by sandGorgon 3y ago
well not so much as the raw data acquisition (scraping and stuff), but really data prep for finetuning. I'm hearing that each model needs it in a different format - chat finetuning data is different from instruct, etc etc
116.
by sandGorgon 3y ago
Quick question - Gorilla paper talks about finetuning for RAG. Do you see this in practice ? can you do finetuning that specifically affects RAG ?
117.
by sandGorgon 3y ago
this is brilliant. could you do a series about how to prepare custom data sets for finetuning. thats the part that a lot of other tutorials skip on. Especially for different goals - like safety, accuracy, etc.
118.
by sandGorgon 3y ago
In EdgeChains, we run models using DeepJavaLibrary (DJL). Preferably only CPU - more focused on edge+embedding usecases
119.
by sandGorgon 3y ago
hi folks, when will you have pgvector as a destination ? we ( https://github.com/arakoodev/edgechains ) work with a lot of enterprises and they would not move away from using redis or pgvector even as their vector store.
120.
by sandGorgon 3y ago
RIP. This makes me really really sad today. I first used vim when i was in college over 25 years back. And it was magical. It was my first tryst with what a text editor should be. My .vimrc is 25 years old :(

More ›