Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
achompas
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
13 ms
·
91.
▲
by
achompas
9y ago
> How about getting the Apple Store people to use their skills on the media store? Because the media store is a legacy product with declining revenue that is already being split into different apps (TV, Apple Music, soon News). I'm
92.
▲
by
achompas
9y ago
I didn't say it would be bad to split things up because of the investment in iTunes since 2001 -- I said it would be hard . I've seen folks confuse iTunes Store with App Store and Apple Store. The split needs to be gradual, wit
93.
▲
by
achompas
9y ago
They've sunk millions of advertising and branding dollars into Apple Music since then, and they've built tons of goodwill from that investment.
94.
▲
by
achompas
9y ago
Let me take a stab: > how come their exclusive TV programming is branded under Apple Music They're using the programming to market Apple Music subscriptions. Apple wants to enroll new users, then up-sell them on new hardware. Ben Th
95.
▲
by
achompas
9y ago
Thanks for sharing. My wife (a medical professional) has commented that I seemingly have trouble breathing through my nose. I'll give strips a shot.
96.
▲
by
achompas
9y ago
This part about sleeping sounds a lot like me. What was the issue, if I may ask?
97.
▲
by
achompas
9y ago
> How much of that is critical domain specific knowledge and how much of that is just general engineering debugging/problem solving experience though? It's both, right? You pick up problem-solving techniques as a researcher or
98.
▲
by
achompas
9y ago
> Aren't you sort of glossing over the fact that he is in high up machine learning position at a company that specializes in recommender systems? Doesn't that by itself increase the likelihood that he deeply understands implici
99.
▲
by
achompas
9y ago
> I would say PHD is sufficient to advance the field. Yep! We’ve now made our way back to my initial point in response to OP. :)
100.
▲
by
achompas
9y ago
Yeah, I expected this reply. The PhD is sufficient but not necessary here, right? A PhD researcher's job description is basically "learn necessary math, become a domain expert, and publish papers advancing that domain." It&#x
101.
▲
by
achompas
9y ago
I think about this constantly. Not to sound like I walked uphill in both directions back in my day or something, but I remember building models in numpy without pandas. It was tedious -- and that's just a nice API wrapping ndarrays!
102.
▲
by
achompas
9y ago
Honestly? The exact problem I'm dealing with at work right now. We're trying to re-write our recommender for artist music stations at iHeartRadio (aka "I'll listen to Drake or Kendrick Lamar's station at the gym tod
103.
▲
by
achompas
9y ago
> both technical, communicative, and even life skills that don't really exist for software devs Not a fan of this "data scientist is a unicorn" style of thinking. The best people in any profession (especially software engi
104.
▲
by
achompas
9y ago
> The obsolete gatekeeping by the AI/ML elites who say "you can't use AI/ML unless you have a PhD/5 years research experience" is one of the things I really hate about the industry. It is absolutely true tha
105.
▲
by
achompas
9y ago
> You could initially present a user some images to assess it's preferences and use some recommender system (collaborative filtering), a la Netflix. You would have to tag the hell out of these photos, right? Disambiguating preferenc
106.
▲
by
achompas
9y ago
You hit the nail on the head. They also reveal this strategy through their data science "thought leadership" posts, which are seemingly meant to appeal to data folks (own your code! add value through modeling! don't worry abo
107.
▲
by
achompas
9y ago
> I never tried myself but it seems quite feasible to build a decent profile of someones taste in fashion from a bit of data. I've thought about this as well. How would you personally try to build this profile for users, then market
108.
▲
by
achompas
9y ago
Seems to me like a person who has made mistakes and learned from them is exactly the type of person whose opinion matters here.
109.
▲
by
achompas
9y ago
> Who is doing this? Why would you sell for USDT instead USD? On many international exchanges, USDT are the easiest way to convert BTC to a (supposedly) USD-backed asset.
110.
▲
by
achompas
9y ago
Important distinction: the notebooks themselves don't do this. The notebook authors, by convention, are supposed to do this. This is no different than engineers who are, by convention, supposed to write bug-free code. Even with thi
111.
▲
by
achompas
9y ago
I might have needed to edit down my comment, but you locked on to the least important part of my argument. If reproducibility is important, like you say, then a notebook is the last thing you want. Your code needs to be tested and designe
112.
▲
by
achompas
9y ago
Notebooks are not much better than copy-pasting from a notepad or editor into an interpreter. They’re great for reports, but dangerous for presenting the illusion of reproducibility. At best you’re constantly restarting your kernel and cl
113.
▲
by
achompas
9y ago
Volume is not the whole market story.
114.
▲
by
achompas
9y ago
Great catch, I missed that. EDIT: Wait, I rushed and misread your comment. Why wouldn't they eliminate these schools if the new schools produced better candidates? I would imagine it was the opposite: remove schools because the new s
115.
▲
by
achompas
9y ago
I wouldn't suggest this as an explanation at all. Another, equally-plausible explanation, is that retail investors are looking for the next BTC before its price pops.
116.
▲
by
achompas
9y ago
> So they did decide to look pass candidate performance. Not quite. Your quote is missing context: they hired the best candidates from the schools they visited. That’s not the set of all schools, though! They tried to correct this by v
117.
▲
by
achompas
9y ago
Interesting feedback on recommenders, thanks for sharing. > Moreover, in all these situations, I feel like it just be improved by asking the user instead. There are thing I'd love recommendations on and I'd be quite willing to
118.
▲
by
achompas
9y ago
I loved working through this years ago. Clear and succinct.
119.
▲
by
achompas
9y ago
I don't agree. If you want to negotiate more PTO at your new job, it's easy to point at your current PTO arrangement and have them match.
120.
▲
by
achompas
9y ago
Can you describe why? Curious for a second opinion.
More ›