Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

HN Office Hours with Kevin and Sam

news.ycombinator.com

Post reply on HN
Print view View original

91–100 of 163 posts

Re: HN Office Hours with Kevin and Sam

#91

Post by trsohmers »

sama wrote: post #74
trsohmers wrote: post #69

Earlier quoted context omitted.

That's pretty awesome that you guys backed a semiconductor company. I didn't apply as I was skeptical you would invest in one... great to see YC expanding so far and fast though. We're looking at having full simulation capabilities (with physical verification in progress) come the end of this year, with a shuttle run (prototype) tape out in Q1 next year. It's about a 3 month turn around, so we are expecting to have o…

This sounds really good. If mobile base stations are the right first customers, go try to get some of them to agree to buy your chip (if you deliver it as planned), and then go try to raise a seed round. After you deliver the evaluation units, try to raise a $10-15MM seed round. I'd focus less on academia and government. They are usually difficult investors and difficult first customers. Non-dilutive funding is nice,…

I've questioned my sanity at many points due to thinking that the "small" HPC market is our best route as that is what none of the big guys are looking at, and they are horribly underserved. Most investors and others we have talked to have said to focus where the current trends/money is (mobile devices themselves, not base stations).

One more question: I'm an engineer at heart, and I want to stay involved in engineering at some level, but don't want to give up control of the direction of the company. Do you think it makes sense/is possible to be CEO and still be involved in doing day to day engineering/design work, and if so any tips on balancing that? If not, at what point does it make sense to bring in a "professional" CEO?

Thanks for the feedback! Would love to grab coffee if you're interested... feel free to send me an email.

Re: HN Office Hours with Kevin and Sam

#92

Post by ph0rque »

sama wrote: post #90
ph0rque wrote: post #78

Earlier quoted context omitted.

Sure, I have no qualms with dropping the open-source label, and opening up the data set. (However, trying to patent the hardware/software might be hard to do, given how much prior art is out there). Do you have any other suggestions on how to build a moat/monopoly?

I'd be tempted to try a Nespresso/Gillette model, where the initial product is cheap and the "refills" are relatively expensive subscriptions, but very high quality and easy to use.

That is a distinct possibility with plant "inserts"[0] for the vegetable beds, one of the planned innovations I've had in the back of my mind. AutoMicroFarm could sell just the empty media inserts for a reasonable price, and also offer pre-planted, ready-to-fruit plants with inserts for a much higher price.

That way, those who are willing to wait weeks/months for their plants to grow from seeds/seedlings can save money, and those who want it now can pay the extra cost.

[0] Something like this, but deeper and made of biodegradable material with mycelium (mushroom roots) growing throughout the material: http://www.koifood.net/contents/media/l_floating_plant_islan...

Re: HN Office Hours with Kevin and Sam

#93

Post by ScottBurson »

trsohmers wrote: post #79
kevin wrote: post #70

Earlier quoted context omitted.

What practical application is your chip best for? What can we do with your chip that we won't be able to do with the current line of chips coming in the next few years if they stay on the current path? The biggest problem for hard technical problems is the storytelling. I went through your site and I couldn't find the sentences I would need to make someone else excited about what you're making in a conversational way…

Hi Kevin, Thanks! I really need to update the website as it is based on our (unsuccessful) fundraising pitch... thankfully, we had some great feedback a couple of months ago on storytelling, and it really improved our pitch and we were able to close a round. I'll make updating our website a priority this weekend. Right now, it is best for signal processing applications and dense matrix (although it is not limited to…

Today, it takes 100 picojoules to do a double precision floating point operation, but 4200 picojoules (42 times!) more energy to move that data from RAM to your registers.

This is a fascinating observation. As one of pg's essays asks, "what do you know that know that no one else knows?" [quoted from memory]. It sounds like this is your answer. Interestingly, it's not literally that no one else knows it -- it's that no one else (AFAIK, of course) has factored it into their designs, as doing so probably requires more-or-less a clean sheet.

Re: HN Office Hours with Kevin and Sam

#94

Post by dzine »

sama wrote: post #80
dzine wrote: post #75

Earlier quoted context omitted.

We picked a reservation concierge service first, as it is a vertical that is immediately testable in SF with easy to understand mechanics for consumers. With the increasing mass-media fueled foodie culture, our hypothesis being that people don’t want the trouble of deciding where to eat and making the reservation - they just wanted a curated experience. And that made it a good candidate for our platform. The other 2…

If the main thing you want to prove out is that you can build a platform where small tasks can be easily farmed out (which again, sounds like a really good idea), I'm very unsure this is the right way to start. It's well established that you can get people to make restaurant reservations for you. The best companies do something either fundamentally new or 10x better than existing solutions. As you think about the fut…

Thanks Sam, really enlightening points.

I realize we might have actually gone about it the wrong way. From a consumer perspective, that’s where we were struggling to find a niche to start off with. Our long term goal is to enable people to buy services from anyone in the world, no matter how small or large (i.e. reservations, travel), and it’s on us to make it actionable by as broad an workforce as possible.

We felt we had to prove out the concept of farming out discrete tasks in a straightforward niche like the concierge-space, as opposed to perhaps a white labeled platform. That probably wasn’t the right approach, and we’d appreciate your thoughts on this.

Re: HN Office Hours with Kevin and Sam

#95

Post by Lukeas14 »

Kluny wrote: post #83
Lukeas14 wrote: post #49

Hey HN, What we do: We're building a community of vehicle data ( http://shadenut.com ). Mechanics and DIYers will be able to look up any piece of information they need to work on their car directly from their phone while still under the hood (ex. torque specs, TSBs, fluid types/capacities, etc). As a developer I've seen the positive effect that StackOverflow has had on our industry as a knowledge base and am trying t…

My suggestion: start with strong subcultures. Become known to drifters as the people who have all the data on 240sx's, or to Jeep people who have specs on every single difference between each model year. Develop a good reputation with each community before moving to the next. Toyodiy is a great resource for me that does what you're doing, but it's not available in handy app form. Basically a comprehensive catalog of…

That's my vision exactly. All the knowledge you would ever need is already out there, probably on page 17 of some forum post. But it'd be a huge to have it all in one place, easily accessible and most importantly, be able to confidently say that it's accurate (the wrong torque on a head bolt could be a $x,000 mistake).

Creating a database schema that reflects the manufacturers model naming and options decisions is indeed a complete pain but we're grinding through that.

I like your idea of finding a small number of the most passionate enthusiast subcultures and growing from there. Would love to hear more about how this app could help you. I'll send you an email.

Re: HN Office Hours with Kevin and Sam

#96

Post by kevin »

Smirnoff wrote: post #12

Hi Kevin and Sam, We are about to finish building a table reservation system (think OpenTable or SeatMe but on steroids). Although it's a "Me Too" product, we will offer features that our competitors can't or won't, e.g. bigger API control for restaurants, various hooks to extend the service such as food delivery, pre-paid reservations, and ticketing for tables to name a few. Essentially, we will offer an iPhone/iPad…

I think you go freemium if the market is large enough to support the idea that if you just convert a small percentage of customers, you'll still be a successful company. If the answer is yes it's large enough, then whatever strategy gets your software in the hands of the most number of users should be the path you follow.

Now, as a startup, I wouldn't start by competing head to head with OpenTable's salesforce going after their customers. That sounds expensive. Also, getting people to switch providers is always way more difficult. Plus, I think people value OT not because of their features but because of the network effect they bring. Don't get me wrong, you'll definitely still be going one-by-one and converting users, but I don't think you should start with those using your competition.

Instead start with new restaurants or those not using anything. To them, you'll feel like a miracle.

The reason you want to maximize distribution of software and engagement (having something that even the free plan is worth talking to people about) is because going freemium is not unlike investing in startups or building up a portfolio.

You invest in your customers in mass (with free features not money) and you hope a small grouping of them will make back that investment many fold.

Like investing, freemium works best as a long term strategy. So when you look at your upgrade triggers, be sure not to cripple that free plan so much that you're optimizing for conversion rate for users as SOON as possible. Convert them when they've grown to be successful. So they can say you were a key part of their success.

I think MailChimp thinks about this really well if you haven't read their ideas already: http://blog.mailchimp.com/going-freemium-one-year-later/

Good luck!

Re: HN Office Hours with Kevin and Sam

#97

Post by sama »

trsohmers wrote: post #91
sama wrote: post #74

Earlier quoted context omitted.

This sounds really good. If mobile base stations are the right first customers, go try to get some of them to agree to buy your chip (if you deliver it as planned), and then go try to raise a seed round. After you deliver the evaluation units, try to raise a $10-15MM seed round. I'd focus less on academia and government. They are usually difficult investors and difficult first customers. Non-dilutive funding is nice,…

I've questioned my sanity at many points due to thinking that the "small" HPC market is our best route as that is what none of the big guys are looking at, and they are horribly underserved. Most investors and others we have talked to have said to focus where the current trends/money is (mobile devices themselves, not base stations). One more question: I'm an engineer at heart, and I want to stay involved in engineer…

Not only do I think it's possible to stay involved in engineering/design work, I think it's optimal, at least for the first few years. The best technical CEOs do this--they're happy to hand off most of the organization to a COO, but not product or engineering. Mark Zuckerberg did this for a long time at Facebook, and probably still does. Every time I meet with Elon Musk I'm struck by how much of the conversation is about engineering details of SpaceX or Tesla.

Re: HN Office Hours with Kevin and Sam

#98

Post by ScottBurson »

trsohmers wrote: post #91
sama wrote: post #74

Earlier quoted context omitted.

This sounds really good. If mobile base stations are the right first customers, go try to get some of them to agree to buy your chip (if you deliver it as planned), and then go try to raise a seed round. After you deliver the evaluation units, try to raise a $10-15MM seed round. I'd focus less on academia and government. They are usually difficult investors and difficult first customers. Non-dilutive funding is nice,…

I've questioned my sanity at many points due to thinking that the "small" HPC market is our best route as that is what none of the big guys are looking at, and they are horribly underserved. Most investors and others we have talked to have said to focus where the current trends/money is (mobile devices themselves, not base stations). One more question: I'm an engineer at heart, and I want to stay involved in engineer…

Ben Horowitz has written some interesting things on this:

http://www.bhorowitz.com/why_we_prefer_founding_ceos/

http://www.bhorowitz.com/why_founders_fail_the_product_ceo_p...

Re: HN Office Hours with Kevin and Sam

#99

Post by trsohmers »

ScottBurson wrote: post #93
trsohmers wrote: post #79

Earlier quoted context omitted.

Hi Kevin, Thanks! I really need to update the website as it is based on our (unsuccessful) fundraising pitch... thankfully, we had some great feedback a couple of months ago on storytelling, and it really improved our pitch and we were able to close a round. I'll make updating our website a priority this weekend. Right now, it is best for signal processing applications and dense matrix (although it is not limited to…

Today, it takes 100 picojoules to do a double precision floating point operation, but 4200 picojoules (42 times!) more energy to move that data from RAM to your registers. This is a fascinating observation. As one of pg's essays asks, "what do you know that know that no one else knows?" [quoted from memory]. It sounds like this is your answer. Interestingly, it's not literally that no one else knows it -- it's that n…

@Scott When designing the architecture over a year ago, we used that fact as the basis of our design paradigm, and while known to some very smart people in the industry, no one has really dared to do any real redesigns in large companies. When I re-read this post (http://blakemasters.com/post/22866240816/peter-thiels-cs183-...) and thought about what our "secret" really was, I decided to include that fact in our presentations and it finally started to "click" for people... technical folk and investors alike.

Re: HN Office Hours with Kevin and Sam

#100

Post by ph0rque »

sama wrote: post #90
ph0rque wrote: post #78

Earlier quoted context omitted.

Sure, I have no qualms with dropping the open-source label, and opening up the data set. (However, trying to patent the hardware/software might be hard to do, given how much prior art is out there). Do you have any other suggestions on how to build a moat/monopoly?

I'd be tempted to try a Nespresso/Gillette model, where the initial product is cheap and the "refills" are relatively expensive subscriptions, but very high quality and easy to use.

Thinking about it more, I could introduce the subscription model by knowing how far along each insert is in the harvest cycle, and offer to ship new inserts right as the existing insert harvest is winding down.
Post reply on HN