ca_tech

10 ms·

31.
by ca_tech 3y ago
I buy a garage door opener. That is the end of my transaction. I buy a connected garage door opener. The provider knows my geolocation, my name, email address, socioeconomic status, even the phone I own. Inferences can be made on activity s
32.
by ca_tech 3y ago
This makes me think of the Alexa skills ecosystem which is full of low quality skills. Many of which have poor data practices abstracted away behind the scenes. How long until "Chat with your favorite character from [Intellectual Prope
33.
by ca_tech 3y ago
I think this is right inline with the utility of multi agent models. Whether distributing tasks to specialized agents trained on domain knowledge or collaborating with context aware agents. I think the context is where we are going to find
34.
by ca_tech 3y ago
Thanks for providing some context Matthew, I specifically went to mastodon to get your take on this.
35.
by ca_tech 3y ago
People may hate on Palantir but sounds like they get it. I agree that the next step is going to be the interaction of multiple models. We have popular models that have been generalized across domains. The next step is models for specific do
36.
by ca_tech 3y ago
I think another wonderful rabbit hole to look at when talking about coffee in a can is the ongoing quest for the self-heating drink can. https://theconversation.com/self-heating-drinks-cans-return-...
37.
by ca_tech 3y ago
An engadget story demonstrates their simulation tool and discusses their "Simulation" venture a little bit more. It does look like the video is animated by AI using pre-generated scenes and characters. https://www.eng
38.
by ca_tech 3y ago
I can see the potential, but I also got lost in the sidebar. The capability to include and query documents and notes is powerful and I liked the way in which you captured the summary of information. Appreciate the complementary credits beca
39.
by ca_tech 3y ago
The article only addresses the website and application. I expect that what we will see is a shift from individual users to enterprise usage. Individuals are not going to subscribe indefinitely, but I expect corporate access to the API will
40.
by ca_tech 3y ago
Agreed, the mind is running a predictive model as we listen to music. The structure and the deviation from structure all play into our mental model of anticipation and emotional response. Some of it is intentional and part of the compositio
41.
by ca_tech 3y ago
May want to add a proxy config to pass proxy information to the curl command. I would expect that some of the more important servers to get alerted about are those NOT able to access the internet directly.
42.
by ca_tech 3y ago
Agreed. I have found it very helpful to not think about how "healthy" something is. Rather think about how the body will use that food as fuel. Andy Galpin has a good video going through the physiology of how the body processes ca
43.
by ca_tech 3y ago
You can work around this by chunking the data using a sliding window. Take the first X amount of sentences that fit into the token length and summarize. Now slide your window of sentences so that you still overlap just a bit with your previ
44.
by ca_tech 3y ago
I had a long history of supporting live events like this. One of the most common phrases we heard from the engineers was: "Well... that didn't act like the model" That is why we test real things in the real world. While the r
45.
by ca_tech 3y ago
Hero Kids is another well regarded kid friendly rpg. Also only relies on 6 sided die and each session is an hour tops. Has been a fun way to spend a rainy day for our kids and they get pretty creative, so much so that the "story"
46.
by ca_tech 3y ago
In this case, I paid for the course which includes a PDF. Using Bing I can query the document and get summary information along with page number references. Only a matter of time until we can do the same query for a site and get video times