boxcardavin

Some of the moves YC is making with their SS feeding the Core program and statements about aiming for 10k startups a year has me worried about how far they can stretch their founder-mentors and partners.

Bias disclosure: I interviewed for S17 but didn't get in.

We'll see, Kalanick has said that without self-driving in Uber's future they are dead.

I think this whole Uber situation is a good reminder to the valley that the "No Assholes" rule is still a good policy.

Multiple stable 2.5amp sources for a Pi and other devices is important and not always reliable with many 12v->5v converters or even with (how I would probably do it) with a power bank that acts as a buffer and is charged via 12v.

It's messy but it works, if that bothers you then you should read about how inefficient compilers can be.

I'd recommend reading The Rise and Fall of American Growth for some useful numbers on this. Between 1870 and 1970 the standard of living increased so dramatically due to a few key inventions that the author calls "onetime" events. Technology since then has led to marginal improvements compared to, say, the widespread adoption of the filtered drinking water networks.

From my experience that is generally because academic = not hacker. By not hacker I mean that they find it hard to slam something together quick and dirty.

Small investors lack the biggest benefit of a good VC – good connections. $120k from 12 non-F&F investors = 12 people to bother you without much benefit.

Kinda, one of the original ideas was to make a beer-dispensing fridge for my neighbor as he is 'slimming' and wanted to have his fridge limit how much beer he drinks. The real genius – dispense the beer INTO the fridge so you only have 1 or 2 cold beers at a time. No one wants to drink warm beer.

Since then it has evolved into a big AI project, will do a Show HN at some point soon.

If this last week's episode does one good thing it will be to convince programmers that it's not that hard to get started doing ML stuff. Not to say that it's all trivial or straightforward, if you look at the linked image below you can see that a simple NN will miss lots of stuff. In this example the image is being examined for hotdogness as a whole, not being segmented or searching for a bounding box of hotdogness.

http://i.imgur.com/LxgowXI.jpg

It tells you a lot about what the person who made the graph thinks about their audience's sophistication.

Offerup has always seemed like a half-baked app that barely works. It has always made me wonder if they do enough business to justify their investments and valuation. I will say that selling from mobile on there is much easier than any Craigslist solution, but the market feels very small even though I am 3 miles from their HQ.

Do it if it's your top priority. Do you have a brand and brand name that is out there and essential to you or your customers right now?

This is super fun and impressive, I'm going to get it going on my machine and start playing immediately.

I don't think the problem here is what the human mind can comprehend because identifying where these methods break down is actually pretty easy from a math perspective, and the breakdown doesn't change as you go up and down with the number of dimensions. What has always surprised me about ML and DL is how far we can stretch simple regression techniques and how useful it is on such a variety of problems.

I agree that it has turned into an NP problem in a lot of cases now but the concepts behind the problem of N-dimensional optimization are pretty well understood.

Eh, it's not really that consequential because anything big will need way more horsepower than you're gonna get on any mobile GPU to be able to done in a reasonable amount of time. We built a CLI tool for our stuff on AWS and our gaming/ML desktop at the office specifically because everyone is on laptops and training or evals are so slow.

Any sense of how much helium will permeate through the grain structure of the drive enclosure during its lifetime?

VCs invest in people they know and people who know people they know. Need-ta fix that first, but those peeps might not be the type of peeps you wanna know.

"The game is the game."

Your SaaS is a subscription to help people build a SaaS? What is the Software component of your service?

Can you describe your training or dataset you used?

The point is that they're not designing their product from any sort of first-principles, it's more a style of design whereby they are using a single hardware widget augmented with a phone. They didn't tailor their product to the problem, which Tesla did.

I'd also love an informed answer about this.

I tried to schedule my wedding for the Saturday before the (Monday) eclipse in my GF's rural Idaho hometown so that my science friends would have a reason to make the 800 mile trip from Seattle. Unfortunately, all hotels in the path of totality have been sold out for months. ALL OF THEM. AirBNBs in an area of Idaho with 3bdrms renting for $650/m are going for $1000/night.

Google's commitment to SDCs has always seemed half hearted so I wonder if this indicates that they are getting more serious about it. Moonshots at X have been getting the axe, the SDC stuff seems to be the only obviously viable one right now.

To the surprise of many folks (myself included) it has turned out that SDC tech is probably for existing manufacturers to develop versus software guys learning how to build cars without the massive supply chains needed to assemble 4000lb widgets. Before anyone points to Tesla, try getting inside a $140k Tesla and then a $140k Mercedes and it will be obvious what advantages there are to having manufacturing experience stretching decades.

I'm always fascinated by people's fascination with smart mirrors. I don't get it myself, but I like their enthusiasm.

You'll probably be ignored here on HN as people will assume you're just trying to growth hack since your title (and post history) suggest you're not a regular.

Aside from that, hello fellow kiwi! Does this violate IGs TOS? It gets pretty gray when you automate interactions on IG or SC, just curious about your experience with being banned or similar. I suggest redoing your page and putting the how it works right at the top, and cut out the marketing-speak. This is a tool so your users will be experts at social and they need to know the meat of what it actually does in as few words as possible. Seriously, cut the marketing-speak even though you're talking to marketers, they'll appreciate the respect for their time.

I have noticed that CS majors tend to be separated into two groups, those who got linear algebra and those who got THROUGH it. Programmers are very logical in their thinking but most are not what I would call traditionally mathematical in their approach to problems.

Fascinating, I'm curious to see if any of the tech press or even mainstream press pick up on this. If they do, will they pixelate the sample images??

What did getting into YC 'liberate' you to do that you later found out was just a mental block?

The (17.6%) increase is unlikely to be something that only one programmer can do, so the market takes over and most devs are already making top 3% income in their countries.

This is a fantastic example and great advice for someone who has accumulated skills but doesn't feel like being an 'expert' in their field.