4 ms·

What did Norvig do at Google, anyone knows?

by didip 5y ago

m0zg 5y ago
Probably the same thing such folks do elsewhere: allowed some very smart people to not step on the rakes quite as often, and be even smarter by providing perspective and advice.
norvig 5y ago
I ran Search for 5 years or so; then ran all of Research for the next 5; then had an increasing smaller portion of a huge growing Research initiative. This past year I enjoyed mentoring startups in ML through the Google for Startups program. But m0gz got it pretty much right.
- lostmsu 5y ago
  Sorry for a snarky comment, but doesn't that cover the time when HN started noticing, that search stopped returning results for the query that was requested, and instead started to be "too clever" about it with no way to override?
  - nostrademons 5y ago
    The time period he's talking about is 2001-2006. HN didn't even exist then. That was when Google was basically like magic and let you find stuff you never knew existed. HN started complaining about Google being too spammy around 2008, and then about results getting too clever around 2012 or so.
    - lostmsu 5y ago
      This was not clear because one of the slices did not mention the span it lasted.
- nothrowaways 5y ago
  Did your job require you to code? How did you manage to keep sharp at coding?
nostrademons 5y ago
He was their first Director of Search Quality, and then switched to being Director of Research. IIRC he had a VP over him when I left (2014), but was still largely calling the shots in Research. Google Research had some very large wins come out of them in the mid/late 00s - their speech recognition and machine translation programs came out of that. AIUI Norvig was also instrumental in Google's research philosophy, which is to embed research teams alongside the products they're developing rather than having a separate research lab that throws papers over the wall for later implementation. Somewhat ironic, given that he ended up heading the dedicated Research department, but Research was viewed as sort of an incubator whose successful projects would be "adopted" by some other product team. He's the reason machine-learning is pervasive at Google and ordinary SWEs use TensorFlow, rather than it being the sole province of Ph.Ds.
vram22 5y ago
I saw that others, including he, replied, but, fun fact, this snippet has been on his site for years: "Note to recruiters: Please don't offer me a job. I already have the best job in the world at the best company in the world. Note to engineers and researchers: see why." (sic)

What did Norvig do at Google, anyone knows?

by didip 5y ago

What did Norvig do at Google, anyone knows?

m0zg 5y ago

Probably the same thing such folks do elsewhere: allowed some very smart people to not step on the rakes quite as often, and be even smarter by providing perspective and advice.

norvig 5y ago

I ran Search for 5 years or so; then ran all of Research for the next 5; then had an increasing smaller portion of a huge growing Research initiative. This past year I enjoyed mentoring startups in ML through the Google for Startups program. But m0gz got it pretty much right.

lostmsu 5y ago
Sorry for a snarky comment, but doesn't that cover the time when HN started noticing, that search stopped returning results for the query that was requested, and instead started to be "too clever" about it with no way to override?
- nostrademons 5y ago
  The time period he's talking about is 2001-2006. HN didn't even exist then. That was when Google was basically like magic and let you find stuff you never knew existed. HN started complaining about Google being too spammy around 2008, and then about results getting too clever around 2012 or so.
  - lostmsu 5y ago
    This was not clear because one of the slices did not mention the span it lasted.
nothrowaways 5y ago
Did your job require you to code? How did you manage to keep sharp at coding?

nostrademons 5y ago

He was their first Director of Search Quality, and then switched to being Director of Research. IIRC he had a VP over him when I left (2014), but was still largely calling the shots in Research. Google Research had some very large wins come out of them in the mid/late 00s - their speech recognition and machine translation programs came out of that. AIUI Norvig was also instrumental in Google's research philosophy, which is to embed research teams alongside the products they're developing rather than having a separate research lab that throws papers over the wall for later implementation. Somewhat ironic, given that he ended up heading the dedicated Research department, but Research was viewed as sort of an incubator whose successful projects would be "adopted" by some other product team. He's the reason machine-learning is pervasive at Google and ordinary SWEs use TensorFlow, rather than it being the sole province of Ph.Ds.

vram22 5y ago

I saw that others, including he, replied, but, fun fact, this snippet has been on his site for years: "Note to recruiters: Please don't offer me a job. I already have the best job in the world at the best company in the world. Note to engineers and researchers: see why." (sic)