foobazgt

18 ms·

181.
by foobazgt 4y ago
That take is ridiculous. The cost to replace a Model Y battery is around $13K (no where near 50% of the cost of the vehicle). More than that, the battery will probably outlast the lifetime of the vehicle, with a design lifespan of approxima
182.
by foobazgt 4y ago
To put this in context, you could have hired a 100 person org at that run rate... from fraudulent texts alone. How does that go unchecked? Mind blown.
183.
by foobazgt 4y ago
$60M in fraudulent texts alone is wild. I couldn't imagine being the director responsible for those costs and not having investigated earlier.
184.
by foobazgt 4y ago
Yes! Well, scrape is a slight exaggeration, but it's more than possible that most of the relevant data came from points guy analysis. I'd suggest reading https://writings.stephenwolfram.com/2023/02/what-i
185.
by foobazgt 4y ago
When you're able to find a prompt for ChatGPT where it doesn't have a lot of data, it becomes immediately and starkly clear how different a next word predictor is from intelligence. This is more difficult than you might naively ex
186.
by foobazgt 4y ago
Browse reddit.com/r/bing for a bit. There's a clear reason why Google has been slow-walking release of its better LLM on one hand, while continuously releasing ML-based improvements across their entire product suite on the ot
187.
by foobazgt 4y ago
Investing in a company and/or integrating with its technology is not the same as acquiring it. For example, it's quite clear that AI is a major part of Google's strategy from the personnel that they have working on it (e.g. J
188.
by foobazgt 4y ago
That doesn't suggest any competence in AI. In contrast, compare to the work that has come directly out of both Google and Meta.
189.
by foobazgt 4y ago
How does Microsoft integrating OpenAI's software demonstrate its own AI prowess?
190.
by foobazgt 4y ago
I signed up to Blind to lurk. I assumed that's why most people do. Like most "anonymous", lightly moderated forums, it's a bit of a cesspool. I wouldn't recommend spending too much time there for the sake of one&#x2
191.
by foobazgt 4y ago
As someone in Georgia who was both one of the first ten employees for a remote FAANG office and also a co-founder of a remote office for an SV unicorn, who got paid more or less what his contemporaries on the west coast did, I'd say yo
192.
by foobazgt 4y ago
Let's agree to come back here in 12 months. If Google is dead, I'll buy you (and myself) several beers.
193.
by foobazgt 4y ago
How is Google sleeping on AI? By all accounts, they have had a better LLM than ChatGPT (Lamda), years in advance of ChatGPT. Google is hampered in that they have a lot to risk. They're a big lawsuit target that can't go around wil
194.
by foobazgt 4y ago
Can't say speak for FAANG, but can confirm for a $100B+ business that analytics was a huge mess. There was constant investment on both the engineering and analytics functions to wrangle a coherent view on top of the underlying operatio
195.
by foobazgt 4y ago
V2X (vehicle to grid, home, or load) is pretty new, but there are some vehicles out there with it. Tesla is not one of them. They seem to be more focused on solving storage with powerwall. I don't know where that's good or bad, bu
196.
by foobazgt 4y ago
I don't know if the specs are public, but from what people have seen, it seems to perform mostly as advertised. Frito Lay is using them in earnest and happy with the results so far: https://insideevs.com/news/63231
197.
by foobazgt 4y ago
Storage in general, and transactional boundaries in particular, are one of the biggest banes of distributed computing. It's still more art than science to determine where and how you slice an entity graph across datastores and services
198.
by foobazgt 4y ago
I just asked Google assistant this: "How many oranges are in a cup of juice?". It responded with: "Two to three oranges", and then followed up with more information why and a link to the source where it derived the infor
199.
by foobazgt 4y ago
There are many non-organizational reasons to use microservices. Here are some examples: As you add more functionality to a system, it gains different workloads, which have different needs (e.g. different storage solutions) and need to scale
200.
by foobazgt 4y ago
"There is a sense where AI/ML amounts to a parasitic use of previously existing culture." Spot on. I've been thinking about this, and you're the first person I've seen mention it. The dystopia is that we get st
201.
by foobazgt 4y ago
I don't think this rule is intended to encompass comments about yourself.
202.
by foobazgt 4y ago
Ok, I agree to disagree then! :)
203.
by foobazgt 4y ago
There are definite cost reductions that can be made, but they pale in comparison to keeping humans out of the loop. For sake of argument, let's pretend you can shave half off the costs. So, now we're talking $8B annually for one s
204.
by foobazgt 4y ago
Savings from scale usually comes from automation, and people are generally looking to talk to humans, not bots. This is the exact reason Google avoids human support in the first place - there are no massive cost savings, because it does not
205.
by foobazgt 4y ago
I think that's a sore underestimation of the support costs Google would have to cover. It looks like support costs average around $4/call. ( https://www.google.com/search?q=how+much+does+the+average+su... ). Google
206.
by foobazgt 4y ago
> (In my experience as a white person it is mostly other white people who think this, but I don't want to generalize.) How to casually stereotype without stereotyping?
207.
by foobazgt 4y ago
You didn't link your reference, so I don't know how those numbers were calculated. However, they are so large they are clearly not maintenance costs, but probably all in costs. That's apples to oranges (Tesla vs Toyota are tw
208.
by foobazgt 4y ago
It's been almost four years for me, and I've had to spend exactly $0 in maintenance on my Model 3. We'll see what the future holds, but not bad so far.
209.
by foobazgt 4y ago
I wasn't privileged to know Chris personally, but I was always keenly waiting to hear his latest achievements and insights. He was clearly talented, passionate, and compassionate. It is so very sad to see him go. Depression is a full b
210.
by foobazgt 4y ago
OP, I've been thinking about starting a column on those kinds of lines - generalized engineering advice. I have 20+ years of experience in the industry and have held both principal engineer and engineering director positions at S&P

More ›