seattleeng

5 ms·

1.
by seattleeng 11mo ago
So they’ll just die when Google has a seamless AI-first mobile experience where consumers just ask their Pixel to do X and it happens. Disruption comes for everyone, hardware isnt a moat
2.
by seattleeng 1y ago
These numbers are certainly wrong lol
3.
by seattleeng 1y ago
It’s more like conditioning the posterior of a response on “Ok, so…” lets the model enter a better latent space for answering logically vs just spitting out a random token.
4.
by seattleeng 2y ago
I think you're almost right I suspect its more a structural issue e.g. because of Big VC (a16z, other firms with dozens or hundreds of investing staff), VCs don't stick around firms long enough for real returns (cash distributed)
5.
by seattleeng 2y ago
literally how openai attracted talent with deepmind as the boogeyman. its a playbook that works
6.
by seattleeng 2y ago
As is usually the case, check the data! A lot of the dataset used has fairly morbid scenarios, so the model is working as expected. All the data was synthetically created with GPT4
7.
by seattleeng 2y ago
Yes, the climate boogeyman shows up as a common anti-tech trope since most people don’t look into the breakdown of emissions. Basically everything in direct consumer control pales in comparison to transit (car/truck/boat/plan
8.
by seattleeng 2y ago
Im reviewing linear algebra now and would also love to know that book!
9.
by seattleeng 2y ago
I actually think there are a few, only some have achieved critical mass with smaller subnetworks. BlueSky and Threads both seem to have attracted non-tech networks although Threads downranked political content, so BlueSky seems like the nat
10.
by seattleeng 2y ago
This is exactly how Ive seen things occur. It makes a lot of sense, given some tools are great new additions (coding especially), but others fall flat (IMO, search)
11.
by seattleeng 2y ago
You can learn things top-down or bottoms-up. I can read & understand most reverse engineering posts like this because I have a strong "bottoms-up" foundation with an EE degree and worked with microcontrollers. But when I read
12.
by seattleeng 2y ago
I think you’re exactly right, it’s a show for insiders to know what these 4 people think so the next time they directly or indirectly encounter them they are known quantities. Its masked as news & informative media but its principally b
13.
by seattleeng 2y ago
Cut 80% of jobs and 80% of revenue That is not what shareholders want, hence why no one is copying it
14.
by seattleeng 2y ago
preserving the vector art after transforming is really cool, anyone know the relevant papers? or was this original research done by Adobe?
15.
by seattleeng 2y ago
Why is working locally important?
16.
by seattleeng 2y ago
A new programming model for distributed computing is desperately needed. Something between a full operational system a la Temporal, but without the extreme operational overhead + a sane cooperative runtime like Golang. I think we're pr
17.
by seattleeng 2y ago
A credit based model with minimum add amounts like that does sound like a great alternative if people prefer usage based pricing. From what Ive seen usage pricing actually drives a lot of anxiety even if the end cost is lower
18.
by seattleeng 2y ago
I get your point but its $3/mo or $18/yr. If a diagram isnt worth 1.80 to you then you probably wouldnt have bought it even with pay as you go
19.
by seattleeng 2y ago
Its cool but this feels really late. I just use Ideogram now. The social features are nice
20.
by seattleeng 2y ago
Was gonna say, the email reads similarly to this famous thread about Java (with the old CTO of MS!): https://www.techemails.com/p/bill-gates-im-literally-losing-... Really makes you think about the structure of mega co
21.
by seattleeng 3y ago
Seems like "open source" as a marketing tactic (or perhaps strategy, if they do continue to release open models) has peaked. I'm not really complaining, we get a lot of stuff for free as engineers (especially software), but i
22.
by seattleeng 3y ago
the majority if the employees at openai joined after chatGPT launched, so it's not like there's some sense of nostalgia or forlorn distress for what they built slowly changing. The stock comp (sorry, "PPUs"... which are
23.
by seattleeng 3y ago
I think there’s an element here thats missing: companies can fund small tiger teams (a la skunkworks) if theres executive buy-in to solve specific targeted problems. A small elite unit of engineers with executive support and external partne
24.
by seattleeng 3y ago
Since you're familiar with this, what was wrong with how Farcaster approached name registration? Signing up or rotating your key is (relatively) cheap on Ethereum mainnet, and client apps could front the cost of signing users up. The u
25.
by seattleeng 4y ago
Also, theres more companies than just FAANG out there that pay comparably, go down levels.fyi and start applying, for anyone who wants to try
26.
by seattleeng 5y ago
WDYM by "practicing only"? Are you unvaccinated (not trying to witch hunt, asking to contextualize) and have you had consistent large social gatherings (reasonably high chance of exposure)? I'm vaccinated but always down to t
27.
by seattleeng 8y ago
OpenTracing doesn’t actually specify when a span is sampled. IIRC Lightstep sends all spans to a service that makes sampling decisions based on moving latency percentiles, while Jaeger works as you described, sampling in the client. Its imp
28.
by seattleeng 8y ago
what is `fd` in this context?
29.
by seattleeng 8y ago
I agree with what you've written, and the phrasing I used was unclear. I was mainly making a note about joining a mid to late stage pre-IPO startup (series B or later) vs joining a large, low growth public company. You can't easil
30.
by seattleeng 8y ago
I agree that there are tiers of companies when it comes to compensation, and this site tends to skew towards recording datas for higher tiers (but many of the companies this site provides levels for do NOT pay as much as FAANG - as more sal

More ›