9 ms·

Another Group of Google Veterans Starts a Self-Driving Technology Company

133 points by artsandsci 10y ago

outside1234 10y ago
Is this the new recruiting for elite recruits? We saw this with the Nest folks, which was really a shell company that hired as many talented folks from Apple as possible, built a product, and then put themselves up for sale in the valley: either back to Apple or to Google. This basically makes joining one of these startups a no-risk opportunity: huge upside if the thing you build actually works out, slide into a bigco job again if it doesn't...
- deepnotderp 10y ago
  What exactly is the downside of that? Is it really that wrong that bright employees who make the actual product have a little more leverage?
  - outside1234 10y ago
    Nope! I just realized that this might be a pattern and not a one-off. :)
    - ryandrake 10y ago
      It's been a pattern for a long time. I remember lot of startups back in the 90s built without any conceivable purpose or goal besides "being bought by Microsoft". I've always thought it would be a great (albeit risky) way for someone who is talented but stuck in a go-nowhere job to get a nice bonus and/or raise: 1. Quit your job doing X 2. Found a company to do some related Y 3. A year layer or so, get purchased by your old company for the technology. It's also an alternative way to get a new job somewhere where you can't seem to get through the interview funnel. I'd imagine that at some Silicon Valley companies it's easier to get acquired than get hired.
  - ben_jones 10y ago
    I see parallels between this and lobbying in the US government where lobbied officials will leave for a private sector position as a lobbyist or executive at say a $BIG_TELCO or $BIG_OIL company. Why shouldn't government professionals use their knowledge and expertise to get more leverage going into the private sector? I think the biggest argument here is that sometimes the end goal becomes not the work that person is currently doing but the payout down the road if their actions line up in a certain way. If your goal at Google is to leave in three years and start a self driving car-company to get acqui-hired, you might for instance spend three years making lists of people to take with you or knowledge that won't violate NDAs or whatever. Though often times that might be acceptable, a.k.a. the 'you don't owe your employer anything' line of logic (which generally I agree with), sometimes it can become morally dubious if for example, it corrupts personal relationships or becomes manipulative. And then finally you get the case of the Waymo guy who may have planned and conspired to steal information on his way out the door, ironically after having already gained multi-generational wealth.
    - SimbaOnSteroids 10y ago
      To your first point, because being part of a government is something you should ideally be doing for a non-monetary altruistic reason, whereas employment at a large corp is clearly about personal gain.
    - sbov 10y ago
      In a vacuum, this isn't a problem. In practice, what you have is people regulating and creating laws for their future employer, and/or legal bribery (knowing you'll get a cushy job for supporting a bill). At this point, you aren't leveraging your knowledge, you are abusing your power.
deepnotderp 10y ago
The bet on maps is interesting. I'm of the personal opinion that in the long run only machine learning can solve autonomous driving, but both MobilEye and now this company both have invested heavily into maps. MobilEye makes sense, they have vested financial reasons that dictate what they do, so they can't turn their back on maps which they've been using for such a long time. On the other hand, starting a company in maps for self driving cars, especially by what appears to be industry veterans, is far more interesting.
- davidlee1435 10y ago
  Fleet management will become a true need, so maybe that will turn out to be their main value prop.
- andreyk 10y ago
  Cruise (now part of GM) also relies heavily on maps, based on a talk I saw their lead CV guy make. It does make a good deal of sense - maps (at least street-maps) are mostly static and they make the problem way easier, so why not use them. In fact, most people these days are implicitly using maps - how often do people drive to new places without GPS navigation? Of course, vision and likely ML are still a necessary part of the equation.
  - deepnotderp 10y ago
    By maps I mean high resolution 3D maps with lidar and stuff as this startup seems to be doing. I don't doubt that 2d maps would be needed like how humans drive with a GPS.
  - potatolicious 10y ago
    > "In fact, most people these days are implicitly using maps - how often do people drive to new places without GPS navigation?" I'd argue also they're using maps in an even more fundamental way - most driving occurs in places where the driver has already been, and relying on a wetware-stored map is a critical part of safe and effective driving. Local knowledge - traffic patterns, weird traffic signal setups, difficult lane changes, etc, is pretty critical to driving well.
malandrew 10y ago
When companies pay each engineer X and earn 10 times X on each engineer, this is inevitable. Near as I can tell, this is a strategy that is helping engineers retain more of the value they are creating. If the difference was more subtle such as 2 to 3 times X, then the incentive to go through this much trouble to retain much of the value you create is much smaller.
- jakelarkin 10y ago
  nit: self-driving talent doesn't currently earn anything, but rather holds some long-term strategic value I would summarize this effect as BigCorp HR being much more disciplined/constrained on individual compensation than the M&A department. Naturally, well-positioned and clever employees will capitalize on this.
  - mmagin 10y ago
    I'm pretty sure there are differing accounting/tax consequences in each sort of spending.
- malandrew 10y ago
  In light of this observation, Google's current lawsuit could be viewed as an attempt to deter employee defection by signaling liabilities for potential acquirers, which will suppress the valuation of any acquisition. Google was already at the center of the high-tech employee wage-fixing earlier this decade, so filing lawsuits to create deterrence wouldn't be that surprising. https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_Litigation https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_L...
  - Qworg 10y ago
    Or the lawsuit could be viewed as an attempt to deter industrial espionage by a competitor within a hot field. Between these, which do you think is more likely?
    - deleted 10y ago
samlittlewood 10y ago
Waiting for one called 'FlashDrive'
return0 10y ago
From their website https://www.deepmap.ai/ https://www.deepmap.ai/ it seems they are focusing on high-definition, centimeter-accuracy maps. Why is that important? It seems with a rough map and training, a robot can find its destination.
- sorenjan 10y ago
  This might have changed since it's been a few years, but I think Google used to use high resolution 3D maps together with LIDAR on their cars to do accurate positioning. If I understood it correctly, it was basically a 3D pattern matching.
- CardenB 10y ago
  Your assumption is wrong. It's still really difficult for vehicles to navigate without high definition maps.
- obmelvin 10y ago
  Traditional maps help a driver get from A to B over a long distance with the user being responsible for the details in between. These types of 3D lidar style maps are meant so the car knows detailed information about roads to inform minute details about the precise section of road to take, etc.
aresant 10y ago
To what degree was Otto a comp for the valuations put on autonomous vehicle startups? Has to be fascinating AND deflating if one of your data points was "holy shit we can make $700m in 6 months if we get a team + MVP out the door!".
PascLeRasc 10y ago
Who wants to start a pool on when Uber will buy it?
steevenwee 10y ago
Lawsuit incoming!
gallerdude 10y ago
This kind of articulates a larger attitude I feel about startups in general - too many people are going for too much. Right now, a bunch of companies are going for too much of the metaphorical self-driving pie. But, there's a lot of niches with less competition - yeah, the slice might be smaller, but I'd rather have a 90% chance at 10% of the pie than a 10% chance at 90% of the pie.
- SimbaOnSteroids 10y ago
  People need to corner some niche of the self driving X market then expand. Going for the whole pie is foolharder unless you already have name recognition.
- malandrew 10y ago
  What niches do you consider unexploited or lacking in competition?
  - gallerdude 10y ago
    I'm not thinking of anything in specific - just in general, more people levitate towards the big things rather than the small things.
- nojvek 10y ago
  Self driving cars will fundamentally change society. By the time we figure out self driving cars, we'll have enough tech for autonomous drones, trucks, planes, mail delivery bots, fork lifts, tractors. I bet whoever capitalizes on this will end up becoming the world's most valuable company. This is a horse-> horseless transformation again.
  - sangnoir 10y ago
    > Self driving cars will fundamentally change society. They said the same about internet shopping way back before the dot com bubble burst. There were hundreds of web-vans, but only one Amazon. I guess everyone thinks of themselves as the Amazon of self driving rather than CueCat[1]. 1. I'm embarrassed to say I badly wanted one. http://content.time.com/time/specials/packages/article/0,28804,1991915_1991909_1991857,00.html http://content.time.com/time/specials/packages/article/0,288...
SEJeff 10y ago
Am I the only person genuinely shocked Google doesn't slap their self driving car engineers with non-competes? I realize that non-competes are illegal by CA law, but perhaps they should do the engineering in a different state to stop hemorrhaging talent that tries to compete with them?
- andrewfong 10y ago
  (1) When most of your existing operations are already in CA, moving to a different state isn't exactly trivial. And if the primary reason you're moving is to enforce a non-compete, I bet it'd make recruitment of the better engineers more difficult. (2) Even if Google did their engineering in another state, let's say the engineer in question then moves to CA. There's a good chance that CA law would still apply.
  - alphonsegaston 10y ago
    #2 Really depends on how bad they want to come after you. There have been cases where non-competes originating out of state have been enforced here in California. I personally wouldn't want to go up against Google's legal team and their war chest.
- malandrew 10y ago
  It's exactly this type of thinking that lead to the decline of Route 128 in the post-war years and the rise of Silicon Valley. If Google were to do this, they'd be cutting off their nose to spite their face. They could not enforce such a contract with existing employees and all new employees would look at such a provision and join an employer who doesn't do this. The truth is that Google lost this market (or never really had it to begin with). At best they will end up like Microsoft, simply extracting rent via their patent licensing program. For more information about Route 128 vs Silicon Valley, I cannot recommend the book Regional Advantage by AnnaLee Saxenian enough. It's a bit dated and academic, but it's fascinating if you want to understand why the tech industry in and around Boston lost to Silicon Valley despite having a huge head start.
- sorenjan 10y ago
npelly 10y ago
Would love to read the 80 page white paper mentioned in the article.
qordoba 10y ago
A bit unrelated to the subject I have a question. As far as I know Google started self-driving tech development much earlier (?) than any other group. However Tesla did it in much quicker timeframe and it is used now as opposed to Google which is still in the lab (?). What took so long for Google? or there is some nuance from their competitors?

133 points by artsandsci 10y ago

Is this the new recruiting for elite recruits? We saw this with the Nest folks, which was really a shell company that hired as many talented folks from Apple as possible, built a product, and then put themselves up for sale in the valley: either back to Apple or to Google. This basically makes joining one of these startups a no-risk opportunity: huge upside if the thing you build actually works out, slide into a bigco job again if it doesn't...

deepnotderp 10y ago
What exactly is the downside of that? Is it really that wrong that bright employees who make the actual product have a little more leverage?
- outside1234 10y ago
  Nope! I just realized that this might be a pattern and not a one-off. :)
  - ryandrake 10y ago
    It's been a pattern for a long time. I remember lot of startups back in the 90s built without any conceivable purpose or goal besides "being bought by Microsoft". I've always thought it would be a great (albeit risky) way for someone who is talented but stuck in a go-nowhere job to get a nice bonus and/or raise: 1. Quit your job doing X 2. Found a company to do some related Y 3. A year layer or so, get purchased by your old company for the technology. It's also an alternative way to get a new job somewhere where you can't seem to get through the interview funnel. I'd imagine that at some Silicon Valley companies it's easier to get acquired than get hired.
- ben_jones 10y ago
  I see parallels between this and lobbying in the US government where lobbied officials will leave for a private sector position as a lobbyist or executive at say a $BIG_TELCO or $BIG_OIL company. Why shouldn't government professionals use their knowledge and expertise to get more leverage going into the private sector? I think the biggest argument here is that sometimes the end goal becomes not the work that person is currently doing but the payout down the road if their actions line up in a certain way. If your goal at Google is to leave in three years and start a self driving car-company to get acqui-hired, you might for instance spend three years making lists of people to take with you or knowledge that won't violate NDAs or whatever. Though often times that might be acceptable, a.k.a. the 'you don't owe your employer anything' line of logic (which generally I agree with), sometimes it can become morally dubious if for example, it corrupts personal relationships or becomes manipulative. And then finally you get the case of the Waymo guy who may have planned and conspired to steal information on his way out the door, ironically after having already gained multi-generational wealth.
  - SimbaOnSteroids 10y ago
    To your first point, because being part of a government is something you should ideally be doing for a non-monetary altruistic reason, whereas employment at a large corp is clearly about personal gain.
  - sbov 10y ago
    In a vacuum, this isn't a problem. In practice, what you have is people regulating and creating laws for their future employer, and/or legal bribery (knowing you'll get a cushy job for supporting a bill). At this point, you aren't leveraging your knowledge, you are abusing your power.

deepnotderp 10y ago

The bet on maps is interesting. I'm of the personal opinion that in the long run only machine learning can solve autonomous driving, but both MobilEye and now this company both have invested heavily into maps. MobilEye makes sense, they have vested financial reasons that dictate what they do, so they can't turn their back on maps which they've been using for such a long time. On the other hand, starting a company in maps for self driving cars, especially by what appears to be industry veterans, is far more interesting.

davidlee1435 10y ago
Fleet management will become a true need, so maybe that will turn out to be their main value prop.
andreyk 10y ago
Cruise (now part of GM) also relies heavily on maps, based on a talk I saw their lead CV guy make. It does make a good deal of sense - maps (at least street-maps) are mostly static and they make the problem way easier, so why not use them. In fact, most people these days are implicitly using maps - how often do people drive to new places without GPS navigation? Of course, vision and likely ML are still a necessary part of the equation.
- deepnotderp 10y ago
  By maps I mean high resolution 3D maps with lidar and stuff as this startup seems to be doing. I don't doubt that 2d maps would be needed like how humans drive with a GPS.
- potatolicious 10y ago
  > "In fact, most people these days are implicitly using maps - how often do people drive to new places without GPS navigation?" I'd argue also they're using maps in an even more fundamental way - most driving occurs in places where the driver has already been, and relying on a wetware-stored map is a critical part of safe and effective driving. Local knowledge - traffic patterns, weird traffic signal setups, difficult lane changes, etc, is pretty critical to driving well.

malandrew 10y ago

When companies pay each engineer X and earn 10 times X on each engineer, this is inevitable. Near as I can tell, this is a strategy that is helping engineers retain more of the value they are creating. If the difference was more subtle such as 2 to 3 times X, then the incentive to go through this much trouble to retain much of the value you create is much smaller.

jakelarkin 10y ago
nit: self-driving talent doesn't currently earn anything, but rather holds some long-term strategic value I would summarize this effect as BigCorp HR being much more disciplined/constrained on individual compensation than the M&A department. Naturally, well-positioned and clever employees will capitalize on this.
- mmagin 10y ago
  I'm pretty sure there are differing accounting/tax consequences in each sort of spending.
malandrew 10y ago
In light of this observation, Google's current lawsuit could be viewed as an attempt to deter employee defection by signaling liabilities for potential acquirers, which will suppress the valuation of any acquisition. Google was already at the center of the high-tech employee wage-fixing earlier this decade, so filing lawsuits to create deterrence wouldn't be that surprising. https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_Litigation https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_L...
- Qworg 10y ago
  Or the lawsuit could be viewed as an attempt to deter industrial espionage by a competitor within a hot field. Between these, which do you think is more likely?
  - deleted 10y ago

samlittlewood 10y ago

Waiting for one called 'FlashDrive'

return0 10y ago

From their website https://www.deepmap.ai/ https://www.deepmap.ai/ it seems they are focusing on high-definition, centimeter-accuracy maps. Why is that important? It seems with a rough map and training, a robot can find its destination.

sorenjan 10y ago
This might have changed since it's been a few years, but I think Google used to use high resolution 3D maps together with LIDAR on their cars to do accurate positioning. If I understood it correctly, it was basically a 3D pattern matching.
CardenB 10y ago
Your assumption is wrong. It's still really difficult for vehicles to navigate without high definition maps.
obmelvin 10y ago
Traditional maps help a driver get from A to B over a long distance with the user being responsible for the details in between. These types of 3D lidar style maps are meant so the car knows detailed information about roads to inform minute details about the precise section of road to take, etc.

aresant 10y ago

To what degree was Otto a comp for the valuations put on autonomous vehicle startups? Has to be fascinating AND deflating if one of your data points was "holy shit we can make $700m in 6 months if we get a team + MVP out the door!".

PascLeRasc 10y ago

Who wants to start a pool on when Uber will buy it?

steevenwee 10y ago

Lawsuit incoming!

gallerdude 10y ago

This kind of articulates a larger attitude I feel about startups in general - too many people are going for too much. Right now, a bunch of companies are going for too much of the metaphorical self-driving pie. But, there's a lot of niches with less competition - yeah, the slice might be smaller, but I'd rather have a 90% chance at 10% of the pie than a 10% chance at 90% of the pie.

SimbaOnSteroids 10y ago
People need to corner some niche of the self driving X market then expand. Going for the whole pie is foolharder unless you already have name recognition.
malandrew 10y ago
What niches do you consider unexploited or lacking in competition?
- gallerdude 10y ago
  I'm not thinking of anything in specific - just in general, more people levitate towards the big things rather than the small things.
nojvek 10y ago
Self driving cars will fundamentally change society. By the time we figure out self driving cars, we'll have enough tech for autonomous drones, trucks, planes, mail delivery bots, fork lifts, tractors. I bet whoever capitalizes on this will end up becoming the world's most valuable company. This is a horse-> horseless transformation again.
- sangnoir 10y ago
  > Self driving cars will fundamentally change society. They said the same about internet shopping way back before the dot com bubble burst. There were hundreds of web-vans, but only one Amazon. I guess everyone thinks of themselves as the Amazon of self driving rather than CueCat[1]. 1. I'm embarrassed to say I badly wanted one. http://content.time.com/time/specials/packages/article/0,28804,1991915_1991909_1991857,00.html http://content.time.com/time/specials/packages/article/0,288...

SEJeff 10y ago

Am I the only person genuinely shocked Google doesn't slap their self driving car engineers with non-competes? I realize that non-competes are illegal by CA law, but perhaps they should do the engineering in a different state to stop hemorrhaging talent that tries to compete with them?

andrewfong 10y ago
(1) When most of your existing operations are already in CA, moving to a different state isn't exactly trivial. And if the primary reason you're moving is to enforce a non-compete, I bet it'd make recruitment of the better engineers more difficult. (2) Even if Google did their engineering in another state, let's say the engineer in question then moves to CA. There's a good chance that CA law would still apply.
- alphonsegaston 10y ago
  #2 Really depends on how bad they want to come after you. There have been cases where non-competes originating out of state have been enforced here in California. I personally wouldn't want to go up against Google's legal team and their war chest.
malandrew 10y ago
It's exactly this type of thinking that lead to the decline of Route 128 in the post-war years and the rise of Silicon Valley. If Google were to do this, they'd be cutting off their nose to spite their face. They could not enforce such a contract with existing employees and all new employees would look at such a provision and join an employer who doesn't do this. The truth is that Google lost this market (or never really had it to begin with). At best they will end up like Microsoft, simply extracting rent via their patent licensing program. For more information about Route 128 vs Silicon Valley, I cannot recommend the book Regional Advantage by AnnaLee Saxenian enough. It's a bit dated and academic, but it's fascinating if you want to understand why the tech industry in and around Boston lost to Silicon Valley despite having a huge head start.
sorenjan 10y ago

npelly 10y ago

Would love to read the 80 page white paper mentioned in the article.

qordoba 10y ago

A bit unrelated to the subject I have a question. As far as I know Google started self-driving tech development much earlier (?) than any other group. However Tesla did it in much quicker timeframe and it is used now as opposed to Google which is still in the lab (?). What took so long for Google? or there is some nuance from their competitors?