WEBVTT Kind: captions Language: en 00:00:15.679 --> 00:00:18.550 align:start position:0% uh<00:00:15.759> welcome<00:00:16.080> everybody<00:00:17.039> this<00:00:17.279> is<00:00:17.440> sf<00:00:17.840> text<00:00:18.400> the 00:00:18.550 --> 00:00:18.560 align:start position:0% uh welcome everybody this is sf text the 00:00:18.560 --> 00:00:20.390 align:start position:0% uh welcome everybody this is sf text the first<00:00:18.880> inaugural<00:00:19.359> meetup<00:00:19.840> uh<00:00:20.080> i'm<00:00:20.320> the 00:00:20.390 --> 00:00:20.400 align:start position:0% first inaugural meetup uh i'm the 00:00:20.400 --> 00:00:22.710 align:start position:0% first inaugural meetup uh i'm the organizer<00:00:20.960> alexis<00:00:21.279> kravrov<00:00:22.160> and<00:00:22.320> we're<00:00:22.560> on 00:00:22.710 --> 00:00:22.720 align:start position:0% organizer alexis kravrov and we're on 00:00:22.720 --> 00:00:24.950 align:start position:0% organizer alexis kravrov and we're on location<00:00:23.279> at<00:00:23.439> wiggling 00:00:24.950 --> 00:00:24.960 align:start position:0% location at wiggling 00:00:24.960 --> 00:00:27.990 align:start position:0% location at wiggling the<00:00:25.119> company<00:00:25.840> which<00:00:26.400> works<00:00:26.880> with<00:00:27.279> a<00:00:27.599> lot<00:00:27.840> of 00:00:27.990 --> 00:00:28.000 align:start position:0% the company which works with a lot of 00:00:28.000 --> 00:00:29.429 align:start position:0% the company which works with a lot of text<00:00:28.320> and<00:00:28.480> uses 00:00:29.429 --> 00:00:29.439 align:start position:0% text and uses 00:00:29.439 --> 00:00:31.269 align:start position:0% text and uses very<00:00:29.679> interesting<00:00:30.080> technology 00:00:31.269 --> 00:00:31.279 align:start position:0% very interesting technology 00:00:31.279 --> 00:00:34.470 align:start position:0% very interesting technology uh<00:00:31.679> we<00:00:31.840> have<00:00:32.320> uh<00:00:32.880> two<00:00:33.200> talks<00:00:33.600> today<00:00:34.239> from 00:00:34.470 --> 00:00:34.480 align:start position:0% uh we have uh two talks today from 00:00:34.480 --> 00:00:37.110 align:start position:0% uh we have uh two talks today from wiggling<00:00:35.040> and<00:00:35.200> from<00:00:35.440> grant<00:00:35.680> and<00:00:35.840> gersol 00:00:37.110 --> 00:00:37.120 align:start position:0% wiggling and from grant and gersol 00:00:37.120 --> 00:00:38.950 align:start position:0% wiggling and from grant and gersol and<00:00:37.280> we<00:00:37.440> are<00:00:37.520> going<00:00:37.600> to<00:00:37.680> talk<00:00:37.920> with<00:00:38.160> adam<00:00:38.640> adam 00:00:38.950 --> 00:00:38.960 align:start position:0% and we are going to talk with adam adam 00:00:38.960 --> 00:00:40.790 align:start position:0% and we are going to talk with adam adam pingel<00:00:39.680> who<00:00:39.840> is<00:00:39.920> our 00:00:40.790 --> 00:00:40.800 align:start position:0% pingel who is our 00:00:40.800 --> 00:00:42.950 align:start position:0% pingel who is our gracious<00:00:41.200> host<00:00:42.000> uh 00:00:42.950 --> 00:00:42.960 align:start position:0% gracious host uh 00:00:42.960 --> 00:00:45.910 align:start position:0% gracious host uh he<00:00:43.120> will<00:00:43.440> first<00:00:43.840> tell<00:00:44.079> us<00:00:44.239> about<00:00:45.120> uh<00:00:45.360> wiggling 00:00:45.910 --> 00:00:45.920 align:start position:0% he will first tell us about uh wiggling 00:00:45.920 --> 00:00:48.549 align:start position:0% he will first tell us about uh wiggling itself<00:00:46.559> so<00:00:46.719> adam<00:00:47.120> what<00:00:47.360> is<00:00:47.440> weak<00:00:47.680> link<00:00:48.079> what 00:00:48.549 --> 00:00:48.559 align:start position:0% itself so adam what is weak link what 00:00:48.559 --> 00:00:49.910 align:start position:0% itself so adam what is weak link what are<00:00:48.719> you<00:00:48.719> guys<00:00:48.960> doing 00:00:49.910 --> 00:00:49.920 align:start position:0% are you guys doing 00:00:49.920 --> 00:00:52.470 align:start position:0% are you guys doing well<00:00:50.320> um<00:00:51.039> we<00:00:51.200> help<00:00:51.440> publishers<00:00:52.000> monetize 00:00:52.470 --> 00:00:52.480 align:start position:0% well um we help publishers monetize 00:00:52.480 --> 00:00:54.790 align:start position:0% well um we help publishers monetize their<00:00:52.640> content<00:00:53.280> in<00:00:53.440> a<00:00:53.520> sentence<00:00:54.399> and<00:00:54.559> that 00:00:54.790 --> 00:00:54.800 align:start position:0% their content in a sentence and that 00:00:54.800 --> 00:00:57.189 align:start position:0% their content in a sentence and that comes<00:00:55.039> in<00:00:55.120> a<00:00:55.199> couple<00:00:55.520> forms<00:00:55.920> we<00:00:56.079> help 00:00:57.189 --> 00:00:57.199 align:start position:0% comes in a couple forms we help 00:00:57.199 --> 00:00:59.270 align:start position:0% comes in a couple forms we help help<00:00:57.440> them<00:00:57.680> affiliate<00:00:58.640> links<00:00:58.960> that<00:00:59.120> they 00:00:59.270 --> 00:00:59.280 align:start position:0% help them affiliate links that they 00:00:59.280 --> 00:01:00.950 align:start position:0% help them affiliate links that they might<00:00:59.520> have<00:00:59.680> already<00:01:00.000> in<00:01:00.079> their<00:01:00.239> content<00:01:00.719> to 00:01:00.950 --> 00:01:00.960 align:start position:0% might have already in their content to 00:01:00.960 --> 00:01:03.670 align:start position:0% might have already in their content to product<00:01:01.199> pages<00:01:02.000> and<00:01:02.079> then<00:01:02.719> we<00:01:02.879> also<00:01:03.520> can 00:01:03.670 --> 00:01:03.680 align:start position:0% product pages and then we also can 00:01:03.680 --> 00:01:05.509 align:start position:0% product pages and then we also can identify<00:01:04.159> product<00:01:04.479> references<00:01:04.960> and<00:01:05.040> text<00:01:05.439> and 00:01:05.509 --> 00:01:05.519 align:start position:0% identify product references and text and 00:01:05.519 --> 00:01:07.990 align:start position:0% identify product references and text and turn<00:01:05.760> those<00:01:06.000> into<00:01:06.320> affiliated<00:01:06.880> links 00:01:07.990 --> 00:01:08.000 align:start position:0% turn those into affiliated links 00:01:08.000 --> 00:01:08.870 align:start position:0% turn those into affiliated links and 00:01:08.870 --> 00:01:08.880 align:start position:0% and 00:01:08.880 --> 00:01:11.590 align:start position:0% and as<00:01:09.200> we're<00:01:09.520> technology<00:01:10.080> people<00:01:10.960> and<00:01:11.200> we<00:01:11.360> like 00:01:11.590 --> 00:01:11.600 align:start position:0% as we're technology people and we like 00:01:11.600 --> 00:01:14.390 align:start position:0% as we're technology people and we like search<00:01:12.080> and<00:01:12.159> nlp<00:01:12.880> and<00:01:12.960> tax<00:01:13.200> money<00:01:14.000> what<00:01:14.240> are 00:01:14.390 --> 00:01:14.400 align:start position:0% search and nlp and tax money what are 00:01:14.400 --> 00:01:16.070 align:start position:0% search and nlp and tax money what are the<00:01:14.799> underlying<00:01:15.200> technologies<00:01:15.760> they<00:01:15.920> use 00:01:16.070 --> 00:01:16.080 align:start position:0% the underlying technologies they use 00:01:16.080 --> 00:01:19.190 align:start position:0% the underlying technologies they use what<00:01:16.320> are<00:01:16.400> the<00:01:16.560> challenges<00:01:17.280> in<00:01:17.520> in<00:01:17.840> doing<00:01:18.159> this 00:01:19.190 --> 00:01:19.200 align:start position:0% what are the challenges in in doing this 00:01:19.200 --> 00:01:20.710 align:start position:0% what are the challenges in in doing this well<00:01:19.600> um 00:01:20.710 --> 00:01:20.720 align:start position:0% well um 00:01:20.720 --> 00:01:22.789 align:start position:0% well um just<00:01:20.880> to<00:01:21.040> speak<00:01:21.280> to<00:01:21.439> the<00:01:21.600> scale<00:01:22.000> of<00:01:22.159> it<00:01:22.320> first 00:01:22.789 --> 00:01:22.799 align:start position:0% just to speak to the scale of it first 00:01:22.799 --> 00:01:24.550 align:start position:0% just to speak to the scale of it first we<00:01:23.119> we<00:01:23.280> work<00:01:23.520> with<00:01:23.759> tens<00:01:24.000> of<00:01:24.159> thousands<00:01:24.479> of 00:01:24.550 --> 00:01:24.560 align:start position:0% we we work with tens of thousands of 00:01:24.560 --> 00:01:26.390 align:start position:0% we we work with tens of thousands of merchants 00:01:26.390 --> 00:01:26.400 align:start position:0% merchants 00:01:26.400 --> 00:01:29.030 align:start position:0% merchants we<00:01:26.560> do<00:01:27.119> many<00:01:27.360> billions<00:01:27.759> of<00:01:27.840> page<00:01:28.159> views 00:01:29.030 --> 00:01:29.040 align:start position:0% we do many billions of page views 00:01:29.040 --> 00:01:30.310 align:start position:0% we do many billions of page views every<00:01:29.280> month 00:01:30.310 --> 00:01:30.320 align:start position:0% every month 00:01:30.320 --> 00:01:32.149 align:start position:0% every month so<00:01:30.560> just<00:01:30.720> scaling 00:01:32.149 --> 00:01:32.159 align:start position:0% so just scaling 00:01:32.159 --> 00:01:34.630 align:start position:0% so just scaling scaling<00:01:32.479> the<00:01:32.640> load<00:01:32.880> is<00:01:33.200> a<00:01:33.280> challenge<00:01:34.000> we<00:01:34.400> run 00:01:34.630 --> 00:01:34.640 align:start position:0% scaling the load is a challenge we run 00:01:34.640 --> 00:01:38.870 align:start position:0% scaling the load is a challenge we run in<00:01:34.880> three<00:01:35.040> different<00:01:35.280> amazon<00:01:35.920> regions 00:01:38.870 --> 00:01:38.880 align:start position:0% 00:01:38.880 --> 00:01:41.749 align:start position:0% databases<00:01:39.439> we<00:01:39.600> use<00:01:39.759> our<00:01:39.920> cassandra<00:01:40.560> mysql<00:01:41.600> uh 00:01:41.749 --> 00:01:41.759 align:start position:0% databases we use our cassandra mysql uh 00:01:41.759 --> 00:01:43.350 align:start position:0% databases we use our cassandra mysql uh we<00:01:41.920> use<00:01:42.079> elasticsearch 00:01:43.350 --> 00:01:43.360 align:start position:0% we use elasticsearch 00:01:43.360 --> 00:01:44.310 align:start position:0% we use elasticsearch um 00:01:44.310 --> 00:01:44.320 align:start position:0% um 00:01:44.320 --> 00:01:46.469 align:start position:0% um so<00:01:44.720> so<00:01:44.880> there's<00:01:45.119> that<00:01:45.280> side<00:01:45.520> of<00:01:45.600> it<00:01:45.759> and<00:01:45.840> then 00:01:46.469 --> 00:01:46.479 align:start position:0% so so there's that side of it and then 00:01:46.479 --> 00:01:47.910 align:start position:0% so so there's that side of it and then on<00:01:46.560> the<00:01:46.720> algorithm<00:01:47.119> side<00:01:47.360> we<00:01:47.520> have<00:01:47.600> a<00:01:47.680> whole 00:01:47.910 --> 00:01:47.920 align:start position:0% on the algorithm side we have a whole 00:01:47.920 --> 00:01:50.069 align:start position:0% on the algorithm side we have a whole data<00:01:48.159> science<00:01:48.479> team<00:01:48.799> and<00:01:49.119> a<00:01:49.200> couple<00:01:49.439> of<00:01:49.600> nlp 00:01:50.069 --> 00:01:50.079 align:start position:0% data science team and a couple of nlp 00:01:50.079 --> 00:01:52.950 align:start position:0% data science team and a couple of nlp experts<00:01:50.960> who<00:01:51.280> uh<00:01:51.759> who<00:01:52.399> tune<00:01:52.640> our<00:01:52.799> name 00:01:52.950 --> 00:01:52.960 align:start position:0% experts who uh who tune our name 00:01:52.960 --> 00:01:54.789 align:start position:0% experts who uh who tune our name identity<00:01:53.280> recognition<00:01:54.000> and<00:01:54.159> our<00:01:54.399> matching 00:01:54.789 --> 00:01:54.799 align:start position:0% identity recognition and our matching 00:01:54.799 --> 00:01:57.270 align:start position:0% identity recognition and our matching technology<00:01:55.840> um<00:01:56.399> to<00:01:56.560> make<00:01:56.720> sure<00:01:56.880> that<00:01:56.960> we<00:01:57.200> are 00:01:57.270 --> 00:01:57.280 align:start position:0% technology um to make sure that we are 00:01:57.280 --> 00:01:58.950 align:start position:0% technology um to make sure that we are linking<00:01:57.600> to<00:01:57.759> high<00:01:58.000> quality<00:01:58.399> products<00:01:58.799> that 00:01:58.950 --> 00:01:58.960 align:start position:0% linking to high quality products that 00:01:58.960 --> 00:02:01.270 align:start position:0% linking to high quality products that are<00:01:59.040> going<00:01:59.200> to<00:01:59.920> yield<00:02:00.479> uh 00:02:01.270 --> 00:02:01.280 align:start position:0% are going to yield uh 00:02:01.280 --> 00:02:04.069 align:start position:0% are going to yield uh the<00:02:01.360> highest<00:02:01.680> possible<00:02:02.560> expected<00:02:03.439> uh<00:02:03.759> revenue 00:02:04.069 --> 00:02:04.079 align:start position:0% the highest possible expected uh revenue 00:02:04.079 --> 00:02:05.749 align:start position:0% the highest possible expected uh revenue for<00:02:04.240> the<00:02:04.320> publishers 00:02:05.749 --> 00:02:05.759 align:start position:0% for the publishers 00:02:05.759 --> 00:02:09.350 align:start position:0% for the publishers uh<00:02:06.079> and<00:02:06.560> i<00:02:06.719> know<00:02:07.360> adam<00:02:08.080> from<00:02:08.399> sf<00:02:08.800> scholar<00:02:09.119> meet 00:02:09.350 --> 00:02:09.360 align:start position:0% uh and i know adam from sf scholar meet 00:02:09.360 --> 00:02:11.430 align:start position:0% uh and i know adam from sf scholar meet up<00:02:09.520> where<00:02:09.759> we've<00:02:09.920> been<00:02:10.160> colleagues<00:02:11.039> in<00:02:11.200> a<00:02:11.280> long 00:02:11.430 --> 00:02:11.440 align:start position:0% up where we've been colleagues in a long 00:02:11.440 --> 00:02:14.790 align:start position:0% up where we've been colleagues in a long time<00:02:11.760> so<00:02:12.400> i<00:02:12.560> wonder<00:02:13.040> what<00:02:13.360> is<00:02:13.599> the<00:02:14.239> rationale 00:02:14.790 --> 00:02:14.800 align:start position:0% time so i wonder what is the rationale 00:02:14.800 --> 00:02:17.350 align:start position:0% time so i wonder what is the rationale behind<00:02:15.360> the<00:02:15.520> choice<00:02:15.760> of<00:02:15.840> scholar<00:02:16.319> and<00:02:16.560> uh<00:02:16.959> its 00:02:17.350 --> 00:02:17.360 align:start position:0% behind the choice of scholar and uh its 00:02:17.360 --> 00:02:20.790 align:start position:0% behind the choice of scholar and uh its uh<00:02:17.680> associated<00:02:18.160> technologies<00:02:18.800> yes<00:02:19.520> um<00:02:20.160> we<00:02:20.480> are 00:02:20.790 --> 00:02:20.800 align:start position:0% uh associated technologies yes um we are 00:02:20.800 --> 00:02:24.229 align:start position:0% uh associated technologies yes um we are historically<00:02:21.440> a<00:02:21.680> java<00:02:22.080> shop<00:02:22.800> um<00:02:23.440> and<00:02:23.680> so<00:02:24.160> i 00:02:24.229 --> 00:02:24.239 align:start position:0% historically a java shop um and so i 00:02:24.239 --> 00:02:25.589 align:start position:0% historically a java shop um and so i think<00:02:24.400> there<00:02:24.640> have<00:02:24.720> been<00:02:24.879> several<00:02:25.200> attempts 00:02:25.589 --> 00:02:25.599 align:start position:0% think there have been several attempts 00:02:25.599 --> 00:02:27.430 align:start position:0% think there have been several attempts over<00:02:25.760> the<00:02:25.840> years<00:02:26.239> to<00:02:26.400> use<00:02:26.560> scala<00:02:26.959> i<00:02:27.120> think<00:02:27.280> the 00:02:27.430 --> 00:02:27.440 align:start position:0% over the years to use scala i think the 00:02:27.440 --> 00:02:29.030 align:start position:0% over the years to use scala i think the gears<00:02:27.840> really<00:02:28.080> started<00:02:28.319> to<00:02:28.400> mesh<00:02:28.640> a<00:02:28.720> couple<00:02:28.959> of 00:02:29.030 --> 00:02:29.040 align:start position:0% gears really started to mesh a couple of 00:02:29.040 --> 00:02:31.110 align:start position:0% gears really started to mesh a couple of years<00:02:29.280> ago<00:02:30.000> we<00:02:30.160> started<00:02:30.480> using<00:02:30.800> it<00:02:30.959> in<00:02:31.040> the 00:02:31.110 --> 00:02:31.120 align:start position:0% years ago we started using it in the 00:02:31.120 --> 00:02:33.910 align:start position:0% years ago we started using it in the feed<00:02:31.440> systems<00:02:32.000> just<00:02:32.160> to<00:02:32.239> kind<00:02:32.400> of<00:02:32.480> get<00:02:32.640> our<00:02:33.200> our 00:02:33.910 --> 00:02:33.920 align:start position:0% feed systems just to kind of get our our 00:02:33.920 --> 00:02:36.949 align:start position:0% feed systems just to kind of get our our you<00:02:34.000> know<00:02:34.239> our<00:02:34.400> feet<00:02:34.640> wet<00:02:35.040> um<00:02:36.080> and 00:02:36.949 --> 00:02:36.959 align:start position:0% you know our feet wet um and 00:02:36.959 --> 00:02:39.830 align:start position:0% you know our feet wet um and it<00:02:37.360> spread<00:02:37.680> from<00:02:37.840> there<00:02:38.160> so<00:02:38.400> now<00:02:38.720> i<00:02:38.800> think 00:02:39.830 --> 00:02:39.840 align:start position:0% it spread from there so now i think 00:02:39.840 --> 00:02:41.830 align:start position:0% it spread from there so now i think close<00:02:40.080> to<00:02:40.319> about<00:02:40.480> a<00:02:40.560> half<00:02:40.879> of<00:02:40.959> our<00:02:41.120> ad<00:02:41.280> hoc 00:02:41.830 --> 00:02:41.840 align:start position:0% close to about a half of our ad hoc 00:02:41.840 --> 00:02:44.630 align:start position:0% close to about a half of our ad hoc analytics<00:02:42.800> uh<00:02:43.440> queries<00:02:44.000> are<00:02:44.319> written<00:02:44.560> in 00:02:44.630 --> 00:02:44.640 align:start position:0% analytics uh queries are written in 00:02:44.640 --> 00:02:47.030 align:start position:0% analytics uh queries are written in scala<00:02:45.120> we<00:02:45.200> do<00:02:45.440> spark 00:02:47.030 --> 00:02:47.040 align:start position:0% scala we do spark 00:02:47.040 --> 00:02:49.190 align:start position:0% scala we do spark we're<00:02:47.440> using<00:02:47.760> akka<00:02:48.239> to<00:02:48.400> do<00:02:48.560> a<00:02:48.560> lot<00:02:48.720> of<00:02:48.879> our<00:02:48.959> feed 00:02:49.190 --> 00:02:49.200 align:start position:0% we're using akka to do a lot of our feed 00:02:49.200 --> 00:02:50.790 align:start position:0% we're using akka to do a lot of our feed processing<00:02:49.680> now 00:02:50.790 --> 00:02:50.800 align:start position:0% processing now 00:02:50.800 --> 00:02:53.509 align:start position:0% processing now we're<00:02:51.040> slowly<00:02:51.519> introducing<00:02:52.080> akka<00:02:52.879> or 00:02:53.509 --> 00:02:53.519 align:start position:0% we're slowly introducing akka or 00:02:53.519 --> 00:02:55.589 align:start position:0% we're slowly introducing akka or play-based<00:02:54.319> uh<00:02:54.560> web<00:02:54.800> services<00:02:55.200> into<00:02:55.440> our 00:02:55.589 --> 00:02:55.599 align:start position:0% play-based uh web services into our 00:02:55.599 --> 00:02:58.550 align:start position:0% play-based uh web services into our infrastructure<00:02:56.400> over<00:02:56.640> time<00:02:57.120> um 00:02:58.550 --> 00:02:58.560 align:start position:0% infrastructure over time um 00:02:58.560 --> 00:03:00.150 align:start position:0% infrastructure over time um so<00:02:58.800> it<00:02:59.120> the<00:02:59.360> the<00:02:59.519> fact<00:02:59.760> that<00:02:59.840> we're<00:02:59.920> coming 00:03:00.150 --> 00:03:00.160 align:start position:0% so it the the fact that we're coming 00:03:00.160 --> 00:03:01.990 align:start position:0% so it the the fact that we're coming from<00:03:00.319> a<00:03:00.400> java<00:03:00.640> background<00:03:01.120> makes<00:03:01.599> to<00:03:01.760> some 00:03:01.990 --> 00:03:02.000 align:start position:0% from a java background makes to some 00:03:02.000 --> 00:03:03.430 align:start position:0% from a java background makes to some extent<00:03:02.239> makes<00:03:02.480> that<00:03:02.560> a<00:03:02.640> lot<00:03:02.800> easier<00:03:03.200> because 00:03:03.430 --> 00:03:03.440 align:start position:0% extent makes that a lot easier because 00:03:03.440 --> 00:03:05.589 align:start position:0% extent makes that a lot easier because it<00:03:03.680> the<00:03:04.000> the<00:03:04.319> migration<00:03:04.720> path<00:03:05.040> is<00:03:05.120> very<00:03:05.280> can<00:03:05.519> be 00:03:05.589 --> 00:03:05.599 align:start position:0% it the the migration path is very can be 00:03:05.599 --> 00:03:07.910 align:start position:0% it the the migration path is very can be very<00:03:05.760> gradual<00:03:06.159> and<00:03:06.239> we<00:03:06.319> can<00:03:06.480> go<00:03:06.640> as<00:03:06.879> at<00:03:07.280> a<00:03:07.680> pace 00:03:07.910 --> 00:03:07.920 align:start position:0% very gradual and we can go as at a pace 00:03:07.920 --> 00:03:10.070 align:start position:0% very gradual and we can go as at a pace that<00:03:08.000> feels<00:03:08.239> comfortable<00:03:08.640> to<00:03:08.800> us<00:03:09.120> um<00:03:09.840> and<00:03:10.000> then 00:03:10.070 --> 00:03:10.080 align:start position:0% that feels comfortable to us um and then 00:03:10.080 --> 00:03:11.670 align:start position:0% that feels comfortable to us um and then we<00:03:10.239> also<00:03:10.480> have<00:03:10.640> a<00:03:10.720> bunch<00:03:10.879> of<00:03:10.959> folks<00:03:11.200> who<00:03:11.360> are<00:03:11.599> a 00:03:11.670 --> 00:03:11.680 align:start position:0% we also have a bunch of folks who are a 00:03:11.680 --> 00:03:13.509 align:start position:0% we also have a bunch of folks who are a little<00:03:11.840> more<00:03:12.000> well-versed<00:03:12.480> in<00:03:12.640> ruby<00:03:13.360> or 00:03:13.509 --> 00:03:13.519 align:start position:0% little more well-versed in ruby or 00:03:13.519 --> 00:03:15.270 align:start position:0% little more well-versed in ruby or python<00:03:14.000> but<00:03:14.159> i<00:03:14.239> think<00:03:14.480> i<00:03:14.560> think<00:03:14.720> they<00:03:14.959> also 00:03:15.270 --> 00:03:15.280 align:start position:0% python but i think i think they also 00:03:15.280 --> 00:03:17.430 align:start position:0% python but i think i think they also find<00:03:15.519> that<00:03:15.680> the<00:03:16.159> functional<00:03:16.480> paradigm<00:03:17.280> is 00:03:17.430 --> 00:03:17.440 align:start position:0% find that the functional paradigm is 00:03:17.440 --> 00:03:19.750 align:start position:0% find that the functional paradigm is pretty<00:03:18.319> pretty<00:03:18.560> natural<00:03:18.959> for<00:03:19.120> them<00:03:19.680> it 00:03:19.750 --> 00:03:19.760 align:start position:0% pretty pretty natural for them it 00:03:19.760 --> 00:03:21.990 align:start position:0% pretty pretty natural for them it doesn't<00:03:20.000> require<00:03:20.400> too<00:03:20.560> much<00:03:21.280> ramp<00:03:21.599> up<00:03:21.680> time<00:03:21.920> to 00:03:21.990 --> 00:03:22.000 align:start position:0% doesn't require too much ramp up time to 00:03:22.000 --> 00:03:23.430 align:start position:0% doesn't require too much ramp up time to get<00:03:22.159> them<00:03:22.319> productive 00:03:23.430 --> 00:03:23.440 align:start position:0% get them productive 00:03:23.440 --> 00:03:25.270 align:start position:0% get them productive cool<00:03:23.920> uh<00:03:24.080> can<00:03:24.159> you<00:03:24.239> talk<00:03:24.480> a<00:03:24.560> lot<00:03:24.800> a<00:03:24.800> little<00:03:25.040> bit 00:03:25.270 --> 00:03:25.280 align:start position:0% cool uh can you talk a lot a little bit 00:03:25.280 --> 00:03:27.910 align:start position:0% cool uh can you talk a lot a little bit about<00:03:25.680> the<00:03:25.920> nlp<00:03:26.400> technologies<00:03:26.879> you're<00:03:27.120> using 00:03:27.910 --> 00:03:27.920 align:start position:0% about the nlp technologies you're using 00:03:27.920 --> 00:03:29.750 align:start position:0% about the nlp technologies you're using uh<00:03:28.400> i<00:03:28.480> know<00:03:28.640> you're<00:03:28.879> related<00:03:29.200> to<00:03:29.280> some<00:03:29.440> of<00:03:29.519> them 00:03:29.750 --> 00:03:29.760 align:start position:0% uh i know you're related to some of them 00:03:29.760 --> 00:03:32.149 align:start position:0% uh i know you're related to some of them i<00:03:29.840> wonder<00:03:30.080> what<00:03:30.319> your<00:03:30.560> experience<00:03:31.599> with<00:03:31.840> them 00:03:32.149 --> 00:03:32.159 align:start position:0% i wonder what your experience with them 00:03:32.159 --> 00:03:34.070 align:start position:0% i wonder what your experience with them you<00:03:32.319> know<00:03:32.799> uh<00:03:33.120> can<00:03:33.280> you<00:03:33.440> recommend<00:03:33.840> some<00:03:34.000> of 00:03:34.070 --> 00:03:34.080 align:start position:0% you know uh can you recommend some of 00:03:34.080 --> 00:03:35.110 align:start position:0% you know uh can you recommend some of them 00:03:35.110 --> 00:03:35.120 align:start position:0% them 00:03:35.120 --> 00:03:37.990 align:start position:0% them so<00:03:35.519> uh<00:03:35.760> yeah<00:03:36.000> we<00:03:36.159> we<00:03:36.319> use<00:03:36.560> mallet<00:03:37.280> um<00:03:37.680> and<00:03:37.840> i'll 00:03:37.990 --> 00:03:38.000 align:start position:0% so uh yeah we we use mallet um and i'll 00:03:38.000 --> 00:03:40.149 align:start position:0% so uh yeah we we use mallet um and i'll have<00:03:38.159> to<00:03:38.319> refer<00:03:38.640> you<00:03:38.720> to<00:03:38.879> our<00:03:39.040> nlp<00:03:39.440> experts<00:03:39.920> for 00:03:40.149 --> 00:03:40.159 align:start position:0% have to refer you to our nlp experts for 00:03:40.159 --> 00:03:42.229 align:start position:0% have to refer you to our nlp experts for for<00:03:40.720> more<00:03:40.879> on<00:03:41.040> that<00:03:41.360> um 00:03:42.229 --> 00:03:42.239 align:start position:0% for more on that um 00:03:42.239 --> 00:03:43.190 align:start position:0% for more on that um mallet 00:03:43.190 --> 00:03:43.200 align:start position:0% mallet 00:03:43.200 --> 00:03:44.949 align:start position:0% mallet we've<00:03:43.519> we've<00:03:43.760> looked<00:03:44.080> at<00:03:44.319> epic<00:03:44.640> we've<00:03:44.799> done 00:03:44.949 --> 00:03:44.959 align:start position:0% we've we've looked at epic we've done 00:03:44.959 --> 00:03:46.070 align:start position:0% we've we've looked at epic we've done some<00:03:45.200> benchmarking<00:03:45.680> with<00:03:45.760> that<00:03:45.840> i<00:03:45.920> think 00:03:46.070 --> 00:03:46.080 align:start position:0% some benchmarking with that i think 00:03:46.080 --> 00:03:47.190 align:start position:0% some benchmarking with that i think katrina<00:03:46.400> is<00:03:46.480> going<00:03:46.560> to<00:03:46.640> talk<00:03:46.879> about<00:03:47.040> that 00:03:47.190 --> 00:03:47.200 align:start position:0% katrina is going to talk about that 00:03:47.200 --> 00:03:49.750 align:start position:0% katrina is going to talk about that tonight<00:03:48.080> um<00:03:48.640> just<00:03:48.799> for<00:03:48.959> the<00:03:49.040> name<00:03:49.200> that<00:03:49.360> any 00:03:49.750 --> 00:03:49.760 align:start position:0% tonight um just for the name that any 00:03:49.760 --> 00:03:51.670 align:start position:0% tonight um just for the name that any recognition<00:03:50.400> we<00:03:50.560> got<00:03:50.799> we<00:03:50.959> did<00:03:51.360> get<00:03:51.519> some 00:03:51.670 --> 00:03:51.680 align:start position:0% recognition we got we did get some 00:03:51.680 --> 00:03:53.990 align:start position:0% recognition we got we did get some higher<00:03:52.000> f1<00:03:52.319> scores<00:03:52.720> but<00:03:53.360> katrina's<00:03:53.840> going<00:03:53.920> to 00:03:53.990 --> 00:03:54.000 align:start position:0% higher f1 scores but katrina's going to 00:03:54.000 --> 00:03:55.990 align:start position:0% higher f1 scores but katrina's going to go<00:03:54.319> into<00:03:54.560> some<00:03:54.799> depth<00:03:55.040> in<00:03:55.120> that<00:03:55.280> later<00:03:55.519> tonight 00:03:55.990 --> 00:03:56.000 align:start position:0% go into some depth in that later tonight 00:03:56.000 --> 00:03:57.509 align:start position:0% go into some depth in that later tonight um 00:03:57.509 --> 00:03:57.519 align:start position:0% um 00:03:57.519 --> 00:03:59.750 align:start position:0% um mallet's<00:03:58.000> natural<00:03:58.400> successor<00:03:58.879> is<00:03:59.040> factory 00:03:59.750 --> 00:03:59.760 align:start position:0% mallet's natural successor is factory 00:03:59.760 --> 00:04:01.830 align:start position:0% mallet's natural successor is factory we've<00:04:00.080> we're<00:04:00.319> aware<00:04:00.640> of<00:04:00.879> it<00:04:01.280> we've<00:04:01.519> looked<00:04:01.760> at 00:04:01.830 --> 00:04:01.840 align:start position:0% we've we're aware of it we've looked at 00:04:01.840 --> 00:04:03.910 align:start position:0% we've we're aware of it we've looked at it<00:04:02.000> we<00:04:02.080> haven't<00:04:02.319> yet<00:04:02.560> benchmarked<00:04:03.120> that<00:04:03.439> but 00:04:03.910 --> 00:04:03.920 align:start position:0% it we haven't yet benchmarked that but 00:04:03.920 --> 00:04:06.869 align:start position:0% it we haven't yet benchmarked that but the<00:04:04.319> epic<00:04:04.720> and<00:04:05.040> factory<00:04:05.439> are<00:04:05.519> the<00:04:05.599> two<00:04:06.000> um 00:04:06.869 --> 00:04:06.879 align:start position:0% the epic and factory are the two um 00:04:06.879 --> 00:04:09.670 align:start position:0% the epic and factory are the two um if<00:04:07.040> we<00:04:07.120> were<00:04:07.280> to<00:04:07.439> migrate<00:04:08.239> beyond<00:04:08.640> mallet<00:04:09.120> uh 00:04:09.670 --> 00:04:09.680 align:start position:0% if we were to migrate beyond mallet uh 00:04:09.680 --> 00:04:11.429 align:start position:0% if we were to migrate beyond mallet uh those<00:04:09.920> are<00:04:10.000> the<00:04:10.239> two<00:04:10.400> most<00:04:10.640> likely<00:04:10.959> libraries 00:04:11.429 --> 00:04:11.439 align:start position:0% those are the two most likely libraries 00:04:11.439 --> 00:04:13.589 align:start position:0% those are the two most likely libraries we'll<00:04:11.920> start<00:04:12.159> using<00:04:12.720> well<00:04:13.040> and<00:04:13.120> so<00:04:13.280> what<00:04:13.519> are 00:04:13.589 --> 00:04:13.599 align:start position:0% we'll start using well and so what are 00:04:13.599 --> 00:04:15.429 align:start position:0% we'll start using well and so what are the<00:04:13.680> main<00:04:13.920> challenges<00:04:14.560> in<00:04:14.799> in<00:04:14.959> working<00:04:15.280> with 00:04:15.429 --> 00:04:15.439 align:start position:0% the main challenges in in working with 00:04:15.439 --> 00:04:17.990 align:start position:0% the main challenges in in working with actual<00:04:15.680> data<00:04:16.479> um<00:04:17.280> and<00:04:17.440> what<00:04:17.680> has<00:04:17.840> been 00:04:17.990 --> 00:04:18.000 align:start position:0% actual data um and what has been 00:04:18.000 --> 00:04:19.349 align:start position:0% actual data um and what has been challenging<00:04:18.400> seeing<00:04:18.639> kind<00:04:18.799> of<00:04:18.880> organizing 00:04:19.349 --> 00:04:19.359 align:start position:0% challenging seeing kind of organizing 00:04:19.359 --> 00:04:21.349 align:start position:0% challenging seeing kind of organizing this<00:04:19.519> information<00:04:20.160> connecting<00:04:20.639> it 00:04:21.349 --> 00:04:21.359 align:start position:0% this information connecting it 00:04:21.359 --> 00:04:23.749 align:start position:0% this information connecting it uh<00:04:21.600> and<00:04:21.840> a<00:04:21.919> little<00:04:22.079> bit<00:04:22.320> about<00:04:22.479> that<00:04:23.040> wow<00:04:23.520> that 00:04:23.749 --> 00:04:23.759 align:start position:0% uh and a little bit about that wow that 00:04:23.759 --> 00:04:25.670 align:start position:0% uh and a little bit about that wow that is<00:04:23.919> a<00:04:24.000> good<00:04:24.160> question 00:04:25.670 --> 00:04:25.680 align:start position:0% is a good question 00:04:25.680 --> 00:04:28.870 align:start position:0% is a good question um 00:04:28.870 --> 00:04:28.880 align:start position:0% 00:04:28.880 --> 00:04:30.710 align:start position:0% gabor<00:04:29.280> is<00:04:29.360> going<00:04:29.440> to<00:04:29.600> speak<00:04:29.840> a<00:04:29.919> lot<00:04:30.160> about 00:04:30.710 --> 00:04:30.720 align:start position:0% gabor is going to speak a lot about 00:04:30.720 --> 00:04:33.670 align:start position:0% gabor is going to speak a lot about creating<00:04:31.199> uh<00:04:31.520> training<00:04:32.000> sets<00:04:32.560> i<00:04:32.800> think 00:04:33.670 --> 00:04:33.680 align:start position:0% creating uh training sets i think 00:04:33.680 --> 00:04:35.670 align:start position:0% creating uh training sets i think uh<00:04:34.400> knowing<00:04:34.800> when<00:04:34.960> you're<00:04:35.120> doing<00:04:35.360> the<00:04:35.440> right 00:04:35.670 --> 00:04:35.680 align:start position:0% uh knowing when you're doing the right 00:04:35.680 --> 00:04:38.150 align:start position:0% uh knowing when you're doing the right thing<00:04:36.160> is<00:04:36.320> challenging<00:04:37.120> so<00:04:37.440> getting<00:04:37.840> all<00:04:38.000> of 00:04:38.150 --> 00:04:38.160 align:start position:0% thing is challenging so getting all of 00:04:38.160 --> 00:04:40.830 align:start position:0% thing is challenging so getting all of the<00:04:38.800> infrastructure<00:04:39.360> in<00:04:39.520> place<00:04:39.919> to 00:04:40.830 --> 00:04:40.840 align:start position:0% the infrastructure in place to 00:04:40.840 --> 00:04:44.310 align:start position:0% the infrastructure in place to um<00:04:42.320> to<00:04:42.479> set<00:04:42.639> up<00:04:42.720> a<00:04:42.880> b<00:04:43.040> tests<00:04:43.759> and<00:04:44.000> make<00:04:44.160> sure 00:04:44.310 --> 00:04:44.320 align:start position:0% um to set up a b tests and make sure 00:04:44.320 --> 00:04:45.990 align:start position:0% um to set up a b tests and make sure that<00:04:44.720> that's<00:04:45.040> all<00:04:45.199> backed<00:04:45.440> up<00:04:45.600> with<00:04:45.840> with 00:04:45.990 --> 00:04:46.000 align:start position:0% that that's all backed up with with 00:04:46.000 --> 00:04:47.990 align:start position:0% that that's all backed up with with solid<00:04:46.400> statistics<00:04:47.360> uh<00:04:47.600> it's<00:04:47.759> very 00:04:47.990 --> 00:04:48.000 align:start position:0% solid statistics uh it's very 00:04:48.000 --> 00:04:49.830 align:start position:0% solid statistics uh it's very challenging<00:04:48.639> um 00:04:49.830 --> 00:04:49.840 align:start position:0% challenging um 00:04:49.840 --> 00:04:51.749 align:start position:0% challenging um the<00:04:50.000> the<00:04:50.240> traffic<00:04:50.560> we<00:04:50.720> see<00:04:50.880> on<00:04:51.040> any<00:04:51.199> given<00:04:51.440> day 00:04:51.749 --> 00:04:51.759 align:start position:0% the the traffic we see on any given day 00:04:51.759 --> 00:04:53.990 align:start position:0% the the traffic we see on any given day it<00:04:52.320> varies 00:04:53.990 --> 00:04:54.000 align:start position:0% it varies 00:04:54.000 --> 00:04:55.430 align:start position:0% it varies it<00:04:54.160> varies<00:04:54.479> a<00:04:54.560> lot<00:04:54.800> the<00:04:54.880> products<00:04:55.199> that<00:04:55.280> people 00:04:55.430 --> 00:04:55.440 align:start position:0% it varies a lot the products that people 00:04:55.440 --> 00:04:57.189 align:start position:0% it varies a lot the products that people are<00:04:55.520> searching<00:04:55.919> for<00:04:56.160> that<00:04:56.320> the<00:04:56.560> seasonality 00:04:57.189 --> 00:04:57.199 align:start position:0% are searching for that the seasonality 00:04:57.199 --> 00:04:59.510 align:start position:0% are searching for that the seasonality of<00:04:57.280> the<00:04:57.360> business<00:04:58.240> um<00:04:58.560> it<00:04:58.639> can<00:04:58.800> be<00:04:59.040> it<00:04:59.199> can<00:04:59.280> be 00:04:59.510 --> 00:04:59.520 align:start position:0% of the business um it can be it can be 00:04:59.520 --> 00:05:01.510 align:start position:0% of the business um it can be it can be difficult<00:04:59.840> to<00:05:00.000> know<00:05:00.639> you<00:05:00.720> know<00:05:00.960> when<00:05:01.199> a<00:05:01.280> change 00:05:01.510 --> 00:05:01.520 align:start position:0% difficult to know you know when a change 00:05:01.520 --> 00:05:02.790 align:start position:0% difficult to know you know when a change you're<00:05:01.600> making<00:05:01.919> is<00:05:02.000> really<00:05:02.160> having<00:05:02.400> desired 00:05:02.790 --> 00:05:02.800 align:start position:0% you're making is really having desired 00:05:02.800 --> 00:05:03.909 align:start position:0% you're making is really having desired effect<00:05:03.199> so 00:05:03.909 --> 00:05:03.919 align:start position:0% effect so 00:05:03.919 --> 00:05:05.510 align:start position:0% effect so um<00:05:04.400> i<00:05:04.639> think 00:05:05.510 --> 00:05:05.520 align:start position:0% um i think 00:05:05.520 --> 00:05:07.270 align:start position:0% um i think that's<00:05:06.000> that's<00:05:06.240> one<00:05:06.479> of<00:05:06.560> the<00:05:06.639> challenges<00:05:07.039> i've 00:05:07.270 --> 00:05:07.280 align:start position:0% that's that's one of the challenges i've 00:05:07.280 --> 00:05:09.909 align:start position:0% that's that's one of the challenges i've observed<00:05:07.759> um<00:05:08.320> but<00:05:08.479> again<00:05:08.720> i<00:05:08.800> think<00:05:09.199> gabor<00:05:09.680> is 00:05:09.909 --> 00:05:09.919 align:start position:0% observed um but again i think gabor is 00:05:09.919 --> 00:05:11.270 align:start position:0% observed um but again i think gabor is is<00:05:10.160> the<00:05:10.240> one<00:05:10.400> who's<00:05:10.639> really<00:05:10.800> leading<00:05:11.039> that 00:05:11.270 --> 00:05:11.280 align:start position:0% is the one who's really leading that 00:05:11.280 --> 00:05:13.590 align:start position:0% is the one who's really leading that charge<00:05:11.600> here<00:05:12.000> um 00:05:13.590 --> 00:05:13.600 align:start position:0% charge here um 00:05:13.600 --> 00:05:16.150 align:start position:0% charge here um uh<00:05:14.000> let's<00:05:14.240> see<00:05:14.400> what<00:05:14.560> else<00:05:14.639> would<00:05:14.880> i<00:05:14.960> say<00:05:15.520> this 00:05:16.150 --> 00:05:16.160 align:start position:0% uh let's see what else would i say this 00:05:16.160 --> 00:05:18.230 align:start position:0% uh let's see what else would i say this scaling<00:05:16.639> search<00:05:17.039> we<00:05:17.199> do<00:05:17.440> we<00:05:17.520> do<00:05:17.680> use<00:05:17.919> search 00:05:18.230 --> 00:05:18.240 align:start position:0% scaling search we do we do use search 00:05:18.240 --> 00:05:19.270 align:start position:0% scaling search we do we do use search we've 00:05:19.270 --> 00:05:19.280 align:start position:0% we've 00:05:19.280 --> 00:05:21.590 align:start position:0% we've gone<00:05:19.680> from<00:05:20.080> using<00:05:20.320> a<00:05:20.479> single<00:05:20.800> box<00:05:21.120> running 00:05:21.590 --> 00:05:21.600 align:start position:0% gone from using a single box running 00:05:21.600 --> 00:05:23.749 align:start position:0% gone from using a single box running lucine<00:05:22.160> embedded<00:05:22.479> in<00:05:22.639> a<00:05:22.720> tomcat<00:05:23.199> container<00:05:23.600> to 00:05:23.749 --> 00:05:23.759 align:start position:0% lucine embedded in a tomcat container to 00:05:23.759 --> 00:05:26.150 align:start position:0% lucine embedded in a tomcat container to running<00:05:24.000> a<00:05:24.080> 30<00:05:24.320> node<00:05:24.560> elastic<00:05:25.039> search<00:05:25.600> cluster 00:05:26.150 --> 00:05:26.160 align:start position:0% running a 30 node elastic search cluster 00:05:26.160 --> 00:05:27.909 align:start position:0% running a 30 node elastic search cluster over<00:05:26.320> the<00:05:26.400> last<00:05:26.639> five<00:05:26.880> years<00:05:27.600> there<00:05:27.680> have<00:05:27.759> been 00:05:27.909 --> 00:05:27.919 align:start position:0% over the last five years there have been 00:05:27.919 --> 00:05:30.629 align:start position:0% over the last five years there have been a<00:05:28.000> couple<00:05:28.240> of<00:05:28.560> points<00:05:28.880> along<00:05:29.120> the<00:05:29.199> way<00:05:29.680> um 00:05:30.629 --> 00:05:30.639 align:start position:0% a couple of points along the way um 00:05:30.639 --> 00:05:31.749 align:start position:0% a couple of points along the way um before 00:05:31.749 --> 00:05:31.759 align:start position:0% before 00:05:31.759 --> 00:05:33.670 align:start position:0% before before<00:05:32.160> there<00:05:32.320> was 00:05:33.670 --> 00:05:33.680 align:start position:0% before there was 00:05:33.680 --> 00:05:36.310 align:start position:0% before there was before<00:05:34.000> there<00:05:34.080> was<00:05:34.720> shard<00:05:35.440> sharding 00:05:36.310 --> 00:05:36.320 align:start position:0% before there was shard sharding 00:05:36.320 --> 00:05:38.469 align:start position:0% before there was shard sharding on<00:05:36.479> solar<00:05:36.960> we<00:05:37.120> had<00:05:37.280> to<00:05:37.360> kind<00:05:37.440> of<00:05:37.520> do<00:05:37.680> our<00:05:37.919> own 00:05:38.469 --> 00:05:38.479 align:start position:0% on solar we had to kind of do our own 00:05:38.479 --> 00:05:40.390 align:start position:0% on solar we had to kind of do our own that<00:05:38.639> was<00:05:38.800> fairly<00:05:39.039> brittle<00:05:39.600> um<00:05:40.000> so<00:05:40.160> about<00:05:40.320> a 00:05:40.390 --> 00:05:40.400 align:start position:0% that was fairly brittle um so about a 00:05:40.400 --> 00:05:42.469 align:start position:0% that was fairly brittle um so about a year<00:05:40.560> ago<00:05:40.720> we<00:05:40.880> moved<00:05:41.039> to<00:05:41.120> elasticsearch 00:05:42.469 --> 00:05:42.479 align:start position:0% year ago we moved to elasticsearch 00:05:42.479 --> 00:05:43.990 align:start position:0% year ago we moved to elasticsearch um 00:05:43.990 --> 00:05:44.000 align:start position:0% um 00:05:44.000 --> 00:05:45.590 align:start position:0% um yeah<00:05:44.160> i<00:05:44.240> would<00:05:44.479> say<00:05:45.039> between<00:05:45.280> the<00:05:45.360> data 00:05:45.590 --> 00:05:45.600 align:start position:0% yeah i would say between the data 00:05:45.600 --> 00:05:47.430 align:start position:0% yeah i would say between the data science<00:05:46.000> science<00:05:46.320> and<00:05:46.400> the<00:05:46.880> and<00:05:46.960> the<00:05:47.039> scaling 00:05:47.430 --> 00:05:47.440 align:start position:0% science science and the and the scaling 00:05:47.440 --> 00:05:48.950 align:start position:0% science science and the and the scaling issues<00:05:48.000> we<00:05:48.080> definitely<00:05:48.400> have<00:05:48.479> our<00:05:48.560> hands<00:05:48.800> full 00:05:48.950 --> 00:05:48.960 align:start position:0% issues we definitely have our hands full 00:05:48.960 --> 00:05:51.029 align:start position:0% issues we definitely have our hands full dealing<00:05:49.280> with<00:05:49.360> this<00:05:49.680> this<00:05:49.919> data 00:05:51.029 --> 00:05:51.039 align:start position:0% dealing with this this data 00:05:51.039 --> 00:05:53.510 align:start position:0% dealing with this this data and<00:05:51.199> also<00:05:51.840> products<00:05:52.479> products<00:05:52.880> themselves<00:05:53.360> as 00:05:53.510 --> 00:05:53.520 align:start position:0% and also products products themselves as 00:05:53.520 --> 00:05:54.469 align:start position:0% and also products products themselves as a<00:05:53.680> rich 00:05:54.469 --> 00:05:54.479 align:start position:0% a rich 00:05:54.479 --> 00:05:56.710 align:start position:0% a rich um 00:05:56.710 --> 00:05:56.720 align:start position:0% um 00:05:56.720 --> 00:05:58.150 align:start position:0% um understanding 00:05:58.150 --> 00:05:58.160 align:start position:0% understanding 00:05:58.160 --> 00:05:59.430 align:start position:0% understanding a<00:05:58.320> product<00:05:58.639> is<00:05:58.720> more<00:05:58.880> than<00:05:59.039> just<00:05:59.280> its 00:05:59.430 --> 00:05:59.440 align:start position:0% a product is more than just its 00:05:59.440 --> 00:06:02.230 align:start position:0% a product is more than just its description<00:06:00.240> there's<00:06:00.880> a<00:06:01.039> rich 00:06:02.230 --> 00:06:02.240 align:start position:0% description there's a rich 00:06:02.240 --> 00:06:03.909 align:start position:0% description there's a rich vocabulary<00:06:02.880> of<00:06:03.039> attributes<00:06:03.440> that<00:06:03.600> products 00:06:03.909 --> 00:06:03.919 align:start position:0% vocabulary of attributes that products 00:06:03.919 --> 00:06:05.749 align:start position:0% vocabulary of attributes that products might<00:06:04.160> have<00:06:04.479> a<00:06:04.720> rich 00:06:05.749 --> 00:06:05.759 align:start position:0% might have a rich 00:06:05.759 --> 00:06:07.830 align:start position:0% might have a rich category 00:06:07.830 --> 00:06:07.840 align:start position:0% category 00:06:07.840 --> 00:06:10.790 align:start position:0% category a<00:06:08.000> taxonomy<00:06:08.560> that<00:06:08.720> you<00:06:08.800> might<00:06:08.960> build<00:06:09.280> uh<00:06:09.840> to<00:06:10.160> to 00:06:10.790 --> 00:06:10.800 align:start position:0% a taxonomy that you might build uh to to 00:06:10.800 --> 00:06:13.749 align:start position:0% a taxonomy that you might build uh to to classify<00:06:11.680> um<00:06:12.000> products<00:06:12.400> so<00:06:12.880> so<00:06:13.039> that<00:06:13.280> work<00:06:13.600> is 00:06:13.749 --> 00:06:13.759 align:start position:0% classify um products so so that work is 00:06:13.759 --> 00:06:15.990 align:start position:0% classify um products so so that work is all<00:06:14.000> ongoing<00:06:14.720> um<00:06:15.039> and<00:06:15.199> still<00:06:15.440> very<00:06:15.600> much<00:06:15.840> we're 00:06:15.990 --> 00:06:16.000 align:start position:0% all ongoing um and still very much we're 00:06:16.000 --> 00:06:17.189 align:start position:0% all ongoing um and still very much we're learning<00:06:16.240> new<00:06:16.400> things<00:06:16.639> about<00:06:16.800> that<00:06:16.960> all<00:06:17.039> the 00:06:17.189 --> 00:06:17.199 align:start position:0% learning new things about that all the 00:06:17.199 --> 00:06:20.070 align:start position:0% learning new things about that all the time<00:06:17.840> great<00:06:18.319> yeah<00:06:18.720> uh<00:06:19.039> so<00:06:19.680> so<00:06:19.919> it's 00:06:20.070 --> 00:06:20.080 align:start position:0% time great yeah uh so so it's 00:06:20.080 --> 00:06:22.230 align:start position:0% time great yeah uh so so it's interesting<00:06:20.479> because<00:06:20.720> you<00:06:20.880> have<00:06:21.280> essentially 00:06:22.230 --> 00:06:22.240 align:start position:0% interesting because you have essentially 00:06:22.240 --> 00:06:24.230 align:start position:0% interesting because you have essentially a<00:06:22.400> team<00:06:22.800> where<00:06:22.960> you<00:06:23.120> have<00:06:23.360> engineers<00:06:23.919> who<00:06:24.080> have 00:06:24.230 --> 00:06:24.240 align:start position:0% a team where you have engineers who have 00:06:24.240 --> 00:06:27.430 align:start position:0% a team where you have engineers who have data<00:06:24.479> scientists<00:06:24.960> who<00:06:25.039> have<00:06:25.280> nlp<00:06:26.160> experts<00:06:27.199> how 00:06:27.430 --> 00:06:27.440 align:start position:0% data scientists who have nlp experts how 00:06:27.440 --> 00:06:30.309 align:start position:0% data scientists who have nlp experts how do<00:06:27.600> you<00:06:27.680> guys<00:06:28.319> balance<00:06:29.199> all<00:06:29.440> of<00:06:29.520> this<00:06:30.000> how<00:06:30.160> do 00:06:30.309 --> 00:06:30.319 align:start position:0% do you guys balance all of this how do 00:06:30.319 --> 00:06:31.350 align:start position:0% do you guys balance all of this how do you<00:06:30.400> kind<00:06:30.639> of 00:06:31.350 --> 00:06:31.360 align:start position:0% you kind of 00:06:31.360 --> 00:06:33.430 align:start position:0% you kind of make<00:06:31.520> sure<00:06:31.759> you<00:06:31.840> have<00:06:32.400> a<00:06:32.639> production<00:06:33.120> code<00:06:33.360> in 00:06:33.430 --> 00:06:33.440 align:start position:0% make sure you have a production code in 00:06:33.440 --> 00:06:35.749 align:start position:0% make sure you have a production code in the<00:06:33.600> end<00:06:33.759> which<00:06:34.000> runs<00:06:34.240> a<00:06:34.319> scale<00:06:35.199> uh<00:06:35.440> can<00:06:35.600> you 00:06:35.749 --> 00:06:35.759 align:start position:0% the end which runs a scale uh can you 00:06:35.759 --> 00:06:38.150 align:start position:0% the end which runs a scale uh can you talk<00:06:35.919> a<00:06:36.000> little<00:06:36.160> bit<00:06:36.400> about<00:06:36.960> uh 00:06:38.150 --> 00:06:38.160 align:start position:0% talk a little bit about uh 00:06:38.160 --> 00:06:41.749 align:start position:0% talk a little bit about uh inter<00:06:39.039> play<00:06:39.440> of<00:06:39.600> these<00:06:40.000> teams<00:06:40.639> and<00:06:40.960> uh<00:06:41.440> how<00:06:41.600> do 00:06:41.749 --> 00:06:41.759 align:start position:0% inter play of these teams and uh how do 00:06:41.759 --> 00:06:43.909 align:start position:0% inter play of these teams and uh how do you<00:06:41.840> get<00:06:42.000> to<00:06:42.160> the<00:06:42.240> actual<00:06:42.479> working<00:06:43.039> yes<00:06:43.680> so<00:06:43.840> we 00:06:43.909 --> 00:06:43.919 align:start position:0% you get to the actual working yes so we 00:06:43.919 --> 00:06:45.749 align:start position:0% you get to the actual working yes so we have<00:06:44.160> we<00:06:44.319> have<00:06:44.479> a<00:06:44.479> little<00:06:44.720> over<00:06:44.880> 20<00:06:45.199> engineers 00:06:45.749 --> 00:06:45.759 align:start position:0% have we have a little over 20 engineers 00:06:45.759 --> 00:06:47.909 align:start position:0% have we have a little over 20 engineers here<00:06:46.240> um<00:06:46.800> most<00:06:47.039> of<00:06:47.120> which<00:06:47.280> are<00:06:47.360> located<00:06:47.680> in<00:06:47.759> san 00:06:47.909 --> 00:06:47.919 align:start position:0% here um most of which are located in san 00:06:47.919 --> 00:06:49.189 align:start position:0% here um most of which are located in san francisco<00:06:48.400> although<00:06:48.639> we<00:06:48.720> do<00:06:48.880> have<00:06:49.039> some 00:06:49.189 --> 00:06:49.199 align:start position:0% francisco although we do have some 00:06:49.199 --> 00:06:51.430 align:start position:0% francisco although we do have some remote<00:06:49.520> offices 00:06:51.430 --> 00:06:51.440 align:start position:0% remote offices 00:06:51.440 --> 00:06:53.189 align:start position:0% remote offices we've<00:06:51.680> actually<00:06:52.000> just<00:06:52.160> recently<00:06:52.880> because 00:06:53.189 --> 00:06:53.199 align:start position:0% we've actually just recently because 00:06:53.199 --> 00:06:54.550 align:start position:0% we've actually just recently because we've<00:06:53.360> been<00:06:53.520> growing<00:06:53.759> so<00:06:53.919> fast<00:06:54.160> split<00:06:54.479> the 00:06:54.550 --> 00:06:54.560 align:start position:0% we've been growing so fast split the 00:06:54.560 --> 00:06:57.510 align:start position:0% we've been growing so fast split the engineering<00:06:54.960> team<00:06:55.520> into<00:06:55.759> two<00:06:56.479> halves<00:06:56.960> and 00:06:57.510 --> 00:06:57.520 align:start position:0% engineering team into two halves and 00:06:57.520 --> 00:06:59.510 align:start position:0% engineering team into two halves and theoretically<00:06:58.160> each<00:06:58.400> team<00:06:58.800> is<00:06:59.039> capable<00:06:59.440> of 00:06:59.510 --> 00:06:59.520 align:start position:0% theoretically each team is capable of 00:06:59.520 --> 00:07:01.670 align:start position:0% theoretically each team is capable of doing<00:06:59.759> any<00:06:59.919> kind<00:07:00.160> of<00:07:00.319> work<00:07:00.800> in<00:07:00.960> practice<00:07:01.440> each 00:07:01.670 --> 00:07:01.680 align:start position:0% doing any kind of work in practice each 00:07:01.680 --> 00:07:04.230 align:start position:0% doing any kind of work in practice each team<00:07:02.080> has<00:07:02.800> has<00:07:03.039> their<00:07:03.199> affinities<00:07:03.840> has<00:07:04.000> their 00:07:04.230 --> 00:07:04.240 align:start position:0% team has has their affinities has their 00:07:04.240 --> 00:07:06.550 align:start position:0% team has has their affinities has their their<00:07:04.479> specializations<00:07:05.520> um 00:07:06.550 --> 00:07:06.560 align:start position:0% their specializations um 00:07:06.560 --> 00:07:07.430 align:start position:0% their specializations um so 00:07:07.430 --> 00:07:07.440 align:start position:0% so 00:07:07.440 --> 00:07:09.830 align:start position:0% so we<00:07:07.599> we<00:07:08.080> found<00:07:08.240> it's<00:07:08.400> beneficial<00:07:09.039> to<00:07:09.280> to<00:07:09.599> keep 00:07:09.830 --> 00:07:09.840 align:start position:0% we we found it's beneficial to to keep 00:07:09.840 --> 00:07:11.110 align:start position:0% we we found it's beneficial to to keep everyone 00:07:11.110 --> 00:07:11.120 align:start position:0% everyone 00:07:11.120 --> 00:07:13.110 align:start position:0% everyone you<00:07:11.280> know<00:07:11.360> well<00:07:11.599> appraised<00:07:12.319> of<00:07:12.479> what<00:07:12.880> what's 00:07:13.110 --> 00:07:13.120 align:start position:0% you know well appraised of what what's 00:07:13.120 --> 00:07:15.510 align:start position:0% you know well appraised of what what's going<00:07:13.360> on<00:07:13.599> across<00:07:14.000> all<00:07:14.160> of<00:07:14.319> engineering<00:07:14.880> so<00:07:15.360> we 00:07:15.510 --> 00:07:15.520 align:start position:0% going on across all of engineering so we 00:07:15.520 --> 00:07:18.150 align:start position:0% going on across all of engineering so we do<00:07:15.680> have<00:07:16.080> data<00:07:16.319> science<00:07:16.720> specific<00:07:17.360> events<00:07:18.000> we 00:07:18.150 --> 00:07:18.160 align:start position:0% do have data science specific events we 00:07:18.160 --> 00:07:19.350 align:start position:0% do have data science specific events we have<00:07:18.240> something<00:07:18.560> called<00:07:18.720> the<00:07:18.800> data<00:07:19.039> science 00:07:19.350 --> 00:07:19.360 align:start position:0% have something called the data science 00:07:19.360 --> 00:07:21.510 align:start position:0% have something called the data science jam<00:07:19.680> every<00:07:20.000> friday<00:07:20.560> at<00:07:20.639> 11<00:07:20.960> o'clock<00:07:21.360> but 00:07:21.510 --> 00:07:21.520 align:start position:0% jam every friday at 11 o'clock but 00:07:21.520 --> 00:07:23.029 align:start position:0% jam every friday at 11 o'clock but that's<00:07:22.000> that's<00:07:22.319> open<00:07:22.639> not<00:07:22.800> only<00:07:22.960> to 00:07:23.029 --> 00:07:23.039 align:start position:0% that's that's open not only to 00:07:23.039 --> 00:07:24.550 align:start position:0% that's that's open not only to engineering<00:07:23.440> but<00:07:23.599> to<00:07:23.759> anyone<00:07:24.000> else<00:07:24.160> in<00:07:24.400> in<00:07:24.479> the 00:07:24.550 --> 00:07:24.560 align:start position:0% engineering but to anyone else in in the 00:07:24.560 --> 00:07:26.950 align:start position:0% engineering but to anyone else in in the company<00:07:24.880> who<00:07:24.960> wants<00:07:25.199> to<00:07:25.440> to<00:07:25.599> check<00:07:25.840> it<00:07:26.000> out 00:07:26.950 --> 00:07:26.960 align:start position:0% company who wants to to check it out 00:07:26.960 --> 00:07:29.350 align:start position:0% company who wants to to check it out it's<00:07:27.120> also<00:07:27.360> a<00:07:27.440> good<00:07:27.599> way<00:07:27.919> for 00:07:29.350 --> 00:07:29.360 align:start position:0% it's also a good way for 00:07:29.360 --> 00:07:31.749 align:start position:0% it's also a good way for for<00:07:29.599> the<00:07:29.680> data<00:07:29.919> scientists<00:07:30.400> to<00:07:30.560> kind<00:07:30.720> of<00:07:30.960> uh 00:07:31.749 --> 00:07:31.759 align:start position:0% for the data scientists to kind of uh 00:07:31.759 --> 00:07:33.430 align:start position:0% for the data scientists to kind of uh you<00:07:31.919> know<00:07:32.240> have<00:07:32.479> a 00:07:33.430 --> 00:07:33.440 align:start position:0% you know have a 00:07:33.440 --> 00:07:35.189 align:start position:0% you know have a produce<00:07:33.759> a<00:07:33.840> deliverable<00:07:34.319> and<00:07:34.560> have<00:07:34.720> a<00:07:34.960> do<00:07:35.120> a 00:07:35.189 --> 00:07:35.199 align:start position:0% produce a deliverable and have a do a 00:07:35.199 --> 00:07:37.670 align:start position:0% produce a deliverable and have a do a talk<00:07:35.520> for<00:07:35.599> a<00:07:35.680> half<00:07:35.919> hour<00:07:36.319> so<00:07:36.800> um<00:07:37.360> we<00:07:37.440> find 00:07:37.670 --> 00:07:37.680 align:start position:0% talk for a half hour so um we find 00:07:37.680 --> 00:07:38.790 align:start position:0% talk for a half hour so um we find that's<00:07:37.840> really<00:07:38.160> what<00:07:38.319> are<00:07:38.400> you<00:07:38.400> guys<00:07:38.639> doing 00:07:38.790 --> 00:07:38.800 align:start position:0% that's really what are you guys doing 00:07:38.800 --> 00:07:41.430 align:start position:0% that's really what are you guys doing these<00:07:38.960> data<00:07:39.199> science<00:07:39.520> jams<00:07:40.319> uh<00:07:40.960> we'll<00:07:41.199> talk 00:07:41.430 --> 00:07:41.440 align:start position:0% these data science jams uh we'll talk 00:07:41.440 --> 00:07:44.550 align:start position:0% these data science jams uh we'll talk about<00:07:42.080> um 00:07:44.550 --> 00:07:44.560 align:start position:0% about um 00:07:44.560 --> 00:07:45.589 align:start position:0% about um we<00:07:44.720> basically<00:07:45.039> do<00:07:45.199> two<00:07:45.360> or<00:07:45.440> three 00:07:45.589 --> 00:07:45.599 align:start position:0% we basically do two or three 00:07:45.599 --> 00:07:47.189 align:start position:0% we basically do two or three presentations<00:07:46.160> over<00:07:46.319> the<00:07:46.400> course<00:07:46.639> of<00:07:46.720> an<00:07:46.880> hour 00:07:47.189 --> 00:07:47.199 align:start position:0% presentations over the course of an hour 00:07:47.199 --> 00:07:50.230 align:start position:0% presentations over the course of an hour and<00:07:47.520> uh<00:07:47.919> the<00:07:48.160> topics<00:07:48.560> can<00:07:48.720> be<00:07:49.360> anything<00:07:49.840> from 00:07:50.230 --> 00:07:50.240 align:start position:0% and uh the topics can be anything from 00:07:50.240 --> 00:07:52.390 align:start position:0% and uh the topics can be anything from restructuring<00:07:50.960> the<00:07:51.039> way<00:07:51.199> we're<00:07:51.440> doing<00:07:52.160> click 00:07:52.390 --> 00:07:52.400 align:start position:0% restructuring the way we're doing click 00:07:52.400 --> 00:07:53.430 align:start position:0% restructuring the way we're doing click ids 00:07:53.430 --> 00:07:53.440 align:start position:0% ids 00:07:53.440 --> 00:07:56.230 align:start position:0% ids um<00:07:53.759> and<00:07:53.840> the<00:07:53.919> math<00:07:54.240> behind<00:07:54.560> that<00:07:54.960> um<00:07:55.520> to 00:07:56.230 --> 00:07:56.240 align:start position:0% um and the math behind that um to 00:07:56.240 --> 00:07:58.790 align:start position:0% um and the math behind that um to to<00:07:56.400> some<00:07:56.639> of<00:07:57.039> gabor's<00:07:57.759> uh 00:07:58.790 --> 00:07:58.800 align:start position:0% to some of gabor's uh 00:07:58.800 --> 00:08:01.189 align:start position:0% to some of gabor's uh research<00:07:59.280> on<00:07:59.599> shallow<00:07:59.919> parsing 00:08:01.189 --> 00:08:01.199 align:start position:0% research on shallow parsing 00:08:01.199 --> 00:08:03.350 align:start position:0% research on shallow parsing it<00:08:01.360> it<00:08:01.440> really<00:08:01.680> runs<00:08:01.919> a<00:08:02.000> pretty<00:08:02.400> wide<00:08:02.639> range<00:08:02.960> of 00:08:03.350 --> 00:08:03.360 align:start position:0% it it really runs a pretty wide range of 00:08:03.360 --> 00:08:04.629 align:start position:0% it it really runs a pretty wide range of topics 00:08:04.629 --> 00:08:04.639 align:start position:0% topics 00:08:04.639 --> 00:08:06.150 align:start position:0% topics statistics<00:08:05.199> behind<00:08:05.520> the<00:08:05.599> kinds<00:08:05.759> of<00:08:05.840> a<00:08:06.000> b 00:08:06.150 --> 00:08:06.160 align:start position:0% statistics behind the kinds of a b 00:08:06.160 --> 00:08:08.309 align:start position:0% statistics behind the kinds of a b testing<00:08:06.400> we're<00:08:06.560> doing<00:08:06.960> a<00:08:07.680> lot<00:08:07.840> of<00:08:07.919> stuff 00:08:08.309 --> 00:08:08.319 align:start position:0% testing we're doing a lot of stuff 00:08:08.319 --> 00:08:11.189 align:start position:0% testing we're doing a lot of stuff awesome<00:08:09.039> do<00:08:09.199> you<00:08:09.280> see<00:08:09.759> uh<00:08:10.400> people<00:08:10.720> learning 00:08:11.189 --> 00:08:11.199 align:start position:0% awesome do you see uh people learning 00:08:11.199 --> 00:08:13.670 align:start position:0% awesome do you see uh people learning and<00:08:11.680> like<00:08:11.919> engineers<00:08:12.879> not<00:08:13.039> yet<00:08:13.199> familiar<00:08:13.599> with 00:08:13.670 --> 00:08:13.680 align:start position:0% and like engineers not yet familiar with 00:08:13.680 --> 00:08:15.189 align:start position:0% and like engineers not yet familiar with these<00:08:13.840> concepts<00:08:14.240> wanting<00:08:14.560> to<00:08:14.720> actually<00:08:15.039> pick 00:08:15.189 --> 00:08:15.199 align:start position:0% these concepts wanting to actually pick 00:08:15.199 --> 00:08:17.189 align:start position:0% these concepts wanting to actually pick it<00:08:15.360> up<00:08:15.680> and<00:08:16.240> do<00:08:16.400> something<00:08:16.639> about<00:08:16.879> it 00:08:17.189 --> 00:08:17.199 align:start position:0% it up and do something about it 00:08:17.199 --> 00:08:19.189 align:start position:0% it up and do something about it absolutely<00:08:17.919> um<00:08:18.319> yeah<00:08:18.560> we<00:08:18.639> actually<00:08:18.879> we<00:08:19.039> just 00:08:19.189 --> 00:08:19.199 align:start position:0% absolutely um yeah we actually we just 00:08:19.199 --> 00:08:21.749 align:start position:0% absolutely um yeah we actually we just had<00:08:19.360> a<00:08:19.520> hack<00:08:19.840> day<00:08:20.319> um<00:08:20.800> which<00:08:21.199> was<00:08:21.440> a<00:08:21.520> good 00:08:21.749 --> 00:08:21.759 align:start position:0% had a hack day um which was a good 00:08:21.759 --> 00:08:23.189 align:start position:0% had a hack day um which was a good chance<00:08:22.160> for<00:08:22.479> some<00:08:22.639> of<00:08:22.639> the<00:08:22.800> engineers<00:08:23.120> who 00:08:23.189 --> 00:08:23.199 align:start position:0% chance for some of the engineers who 00:08:23.199 --> 00:08:25.029 align:start position:0% chance for some of the engineers who don't<00:08:23.360> get<00:08:23.840> don't<00:08:24.080> touch<00:08:24.319> that<00:08:24.479> stuff<00:08:24.720> day<00:08:24.879> to 00:08:25.029 --> 00:08:25.039 align:start position:0% don't get don't touch that stuff day to 00:08:25.039 --> 00:08:27.350 align:start position:0% don't get don't touch that stuff day to day<00:08:25.840> to<00:08:26.080> interact<00:08:26.479> not<00:08:26.720> only<00:08:26.879> with<00:08:27.039> the<00:08:27.120> data 00:08:27.350 --> 00:08:27.360 align:start position:0% day to interact not only with the data 00:08:27.360 --> 00:08:29.830 align:start position:0% day to interact not only with the data scientists<00:08:27.840> but<00:08:28.160> but<00:08:28.639> folks<00:08:28.960> in<00:08:29.039> marketing<00:08:29.599> or 00:08:29.830 --> 00:08:29.840 align:start position:0% scientists but but folks in marketing or 00:08:29.840 --> 00:08:33.269 align:start position:0% scientists but but folks in marketing or operations<00:08:30.639> we<00:08:30.960> had<00:08:31.120> a<00:08:31.759> pan<00:08:32.159> company<00:08:32.880> hack<00:08:33.120> day 00:08:33.269 --> 00:08:33.279 align:start position:0% operations we had a pan company hack day 00:08:33.279 --> 00:08:35.110 align:start position:0% operations we had a pan company hack day so<00:08:33.440> that<00:08:33.599> was<00:08:33.680> a<00:08:33.839> really<00:08:34.080> great<00:08:34.320> opportunity 00:08:35.110 --> 00:08:35.120 align:start position:0% so that was a really great opportunity 00:08:35.120 --> 00:08:35.909 align:start position:0% so that was a really great opportunity um 00:08:35.909 --> 00:08:35.919 align:start position:0% um 00:08:35.919 --> 00:08:37.110 align:start position:0% um to<00:08:36.159> to 00:08:37.110 --> 00:08:37.120 align:start position:0% to to 00:08:37.120 --> 00:08:39.029 align:start position:0% to to give<00:08:37.360> folks<00:08:37.680> enough<00:08:38.159> give<00:08:38.320> folks<00:08:38.560> a<00:08:38.640> chance<00:08:38.959> to 00:08:39.029 --> 00:08:39.039 align:start position:0% give folks enough give folks a chance to 00:08:39.039 --> 00:08:40.630 align:start position:0% give folks enough give folks a chance to work<00:08:39.200> with<00:08:39.360> stuff<00:08:39.519> that<00:08:39.599> they<00:08:39.760> don't<00:08:39.919> normally 00:08:40.630 --> 00:08:40.640 align:start position:0% work with stuff that they don't normally 00:08:40.640 --> 00:08:42.630 align:start position:0% work with stuff that they don't normally uh<00:08:40.800> get<00:08:40.959> to<00:08:41.039> work<00:08:41.200> with<00:08:41.360> day<00:08:41.519> to<00:08:41.599> day<00:08:42.159> excellent 00:08:42.630 --> 00:08:42.640 align:start position:0% uh get to work with day to day excellent 00:08:42.640 --> 00:08:45.190 align:start position:0% uh get to work with day to day excellent excellent<00:08:43.279> uh<00:08:43.599> so<00:08:43.760> grant<00:08:44.159> just<00:08:44.480> join<00:08:44.800> us<00:08:45.040> and 00:08:45.190 --> 00:08:45.200 align:start position:0% excellent uh so grant just join us and 00:08:45.200 --> 00:08:47.030 align:start position:0% excellent uh so grant just join us and you<00:08:45.360> will<00:08:45.440> have<00:08:45.600> to<00:08:45.680> explain<00:08:46.000> to<00:08:46.160> him<00:08:46.640> why<00:08:46.880> you 00:08:47.030 --> 00:08:47.040 align:start position:0% you will have to explain to him why you 00:08:47.040 --> 00:08:48.790 align:start position:0% you will have to explain to him why you picked<00:08:47.279> classic<00:08:47.600> search<00:08:48.000> over<00:08:48.399> i<00:08:48.480> know<00:08:48.720> i'm 00:08:48.790 --> 00:08:48.800 align:start position:0% picked classic search over i know i'm 00:08:48.800 --> 00:08:51.190 align:start position:0% picked classic search over i know i'm gonna<00:08:49.040> have<00:08:49.120> to<00:08:49.200> give<00:08:49.360> him<00:08:49.440> a<00:08:49.519> heads<00:08:49.760> up 00:08:51.190 --> 00:08:51.200 align:start position:0% gonna have to give him a heads up 00:08:51.200 --> 00:08:53.110 align:start position:0% gonna have to give him a heads up yeah<00:08:51.519> so<00:08:51.680> this<00:08:51.839> is<00:08:51.920> great<00:08:52.080> so<00:08:52.320> but<00:08:52.640> since<00:08:52.880> this 00:08:53.110 --> 00:08:53.120 align:start position:0% yeah so this is great so but since this 00:08:53.120 --> 00:08:56.630 align:start position:0% yeah so this is great so but since this is<00:08:53.200> the<00:08:53.360> first<00:08:53.760> sf<00:08:54.080> text<00:08:54.320> method<00:08:55.200> and<00:08:55.600> uh<00:08:56.320> and 00:08:56.630 --> 00:08:56.640 align:start position:0% is the first sf text method and uh and 00:08:56.640 --> 00:08:58.870 align:start position:0% is the first sf text method and uh and we<00:08:56.720> are<00:08:56.880> set<00:08:57.040> in<00:08:57.120> the<00:08:57.200> direction<00:08:58.000> and 00:08:58.870 --> 00:08:58.880 align:start position:0% we are set in the direction and 00:08:58.880 --> 00:09:01.910 align:start position:0% we are set in the direction and if<00:08:59.040> we<00:08:59.200> link<00:08:59.600> has<00:08:59.839> taxes<00:09:00.240> at<00:09:00.399> score<00:09:00.800> i<00:09:00.959> wonder 00:09:01.910 --> 00:09:01.920 align:start position:0% if we link has taxes at score i wonder 00:09:01.920 --> 00:09:05.110 align:start position:0% if we link has taxes at score i wonder what<00:09:02.240> would<00:09:02.640> you<00:09:03.040> as<00:09:03.279> the<00:09:03.440> host<00:09:03.760> and<00:09:03.920> engineer 00:09:05.110 --> 00:09:05.120 align:start position:0% what would you as the host and engineer 00:09:05.120 --> 00:09:07.509 align:start position:0% what would you as the host and engineer uh<00:09:05.440> running<00:09:05.760> a<00:09:05.839> team<00:09:06.160> here<00:09:06.880> what<00:09:07.120> would<00:09:07.360> you 00:09:07.509 --> 00:09:07.519 align:start position:0% uh running a team here what would you 00:09:07.519 --> 00:09:09.590 align:start position:0% uh running a team here what would you like<00:09:07.920> to<00:09:08.640> kind<00:09:08.800> of 00:09:09.590 --> 00:09:09.600 align:start position:0% like to kind of 00:09:09.600 --> 00:09:11.750 align:start position:0% like to kind of get<00:09:09.839> from<00:09:10.080> these<00:09:10.320> meetups<00:09:11.040> and<00:09:11.279> going<00:09:11.600> to<00:09:11.680> the 00:09:11.750 --> 00:09:11.760 align:start position:0% get from these meetups and going to the 00:09:11.760 --> 00:09:13.430 align:start position:0% get from these meetups and going to the future<00:09:12.240> how<00:09:12.399> do<00:09:12.560> you<00:09:12.720> see<00:09:12.959> the<00:09:13.200> most 00:09:13.430 --> 00:09:13.440 align:start position:0% future how do you see the most 00:09:13.440 --> 00:09:14.790 align:start position:0% future how do you see the most beneficial 00:09:14.790 --> 00:09:14.800 align:start position:0% beneficial 00:09:14.800 --> 00:09:17.750 align:start position:0% beneficial uh<00:09:15.519> composition<00:09:16.160> of<00:09:16.240> these<00:09:16.399> meet<00:09:16.640> ups<00:09:17.040> yeah 00:09:17.750 --> 00:09:17.760 align:start position:0% uh composition of these meet ups yeah 00:09:17.760 --> 00:09:19.590 align:start position:0% uh composition of these meet ups yeah well<00:09:17.920> i<00:09:18.240> think<00:09:18.480> uh 00:09:19.590 --> 00:09:19.600 align:start position:0% well i think uh 00:09:19.600 --> 00:09:20.710 align:start position:0% well i think uh a<00:09:19.680> lot<00:09:19.839> of<00:09:19.920> the<00:09:20.000> meetups<00:09:20.320> here<00:09:20.480> in<00:09:20.560> san 00:09:20.710 --> 00:09:20.720 align:start position:0% a lot of the meetups here in san 00:09:20.720 --> 00:09:22.630 align:start position:0% a lot of the meetups here in san francisco<00:09:21.040> tend<00:09:21.200> to<00:09:21.279> be<00:09:21.440> technology<00:09:22.160> oriented 00:09:22.630 --> 00:09:22.640 align:start position:0% francisco tend to be technology oriented 00:09:22.640 --> 00:09:23.990 align:start position:0% francisco tend to be technology oriented this<00:09:22.800> is<00:09:22.880> something<00:09:23.200> that<00:09:23.360> katrine<00:09:23.760> has 00:09:23.990 --> 00:09:24.000 align:start position:0% this is something that katrine has 00:09:24.000 --> 00:09:26.070 align:start position:0% this is something that katrine has comments<00:09:24.399> to<00:09:24.480> me<00:09:24.959> about<00:09:25.360> from 00:09:26.070 --> 00:09:26.080 align:start position:0% comments to me about from 00:09:26.080 --> 00:09:27.670 align:start position:0% comments to me about from from<00:09:26.399> joining<00:09:26.720> us 00:09:27.670 --> 00:09:27.680 align:start position:0% from joining us 00:09:27.680 --> 00:09:29.670 align:start position:0% from joining us almost<00:09:27.920> a<00:09:28.000> year<00:09:28.160> ago<00:09:28.399> she<00:09:28.560> noticed<00:09:28.880> this<00:09:29.040> so<00:09:29.600> i 00:09:29.670 --> 00:09:29.680 align:start position:0% almost a year ago she noticed this so i 00:09:29.680 --> 00:09:32.070 align:start position:0% almost a year ago she noticed this so i think<00:09:29.920> having<00:09:30.640> having<00:09:30.959> a<00:09:31.360> community<00:09:31.839> built 00:09:32.070 --> 00:09:32.080 align:start position:0% think having having a community built 00:09:32.080 --> 00:09:33.670 align:start position:0% think having having a community built around<00:09:32.399> a<00:09:32.480> problem<00:09:32.800> domain<00:09:33.200> more<00:09:33.360> than<00:09:33.519> a 00:09:33.670 --> 00:09:33.680 align:start position:0% around a problem domain more than a 00:09:33.680 --> 00:09:35.670 align:start position:0% around a problem domain more than a technology<00:09:34.240> would<00:09:34.399> be<00:09:34.560> really<00:09:35.120> really<00:09:35.360> useful 00:09:35.670 --> 00:09:35.680 align:start position:0% technology would be really really useful 00:09:35.680 --> 00:09:37.829 align:start position:0% technology would be really really useful to<00:09:35.839> everyone<00:09:36.160> here<00:09:36.320> at<00:09:36.399> big<00:09:36.560> link 00:09:37.829 --> 00:09:37.839 align:start position:0% to everyone here at big link 00:09:37.839 --> 00:09:39.910 align:start position:0% to everyone here at big link and<00:09:38.480> of<00:09:38.560> course<00:09:38.800> just<00:09:38.959> having<00:09:39.279> people<00:09:39.519> over<00:09:39.760> in 00:09:39.910 --> 00:09:39.920 align:start position:0% and of course just having people over in 00:09:39.920 --> 00:09:42.470 align:start position:0% and of course just having people over in person<00:09:40.240> meeting<00:09:40.560> them<00:09:40.959> face<00:09:41.200> to<00:09:41.360> face<00:09:42.160> a<00:09:42.320> lot 00:09:42.470 --> 00:09:42.480 align:start position:0% person meeting them face to face a lot 00:09:42.480 --> 00:09:46.070 align:start position:0% person meeting them face to face a lot of<00:09:42.640> us<00:09:43.200> see<00:09:43.600> these<00:09:43.839> people<00:09:44.640> in<00:09:45.040> in<00:09:45.360> get<00:09:45.600> commits 00:09:46.070 --> 00:09:46.080 align:start position:0% of us see these people in in get commits 00:09:46.080 --> 00:09:48.230 align:start position:0% of us see these people in in get commits or<00:09:46.320> in<00:09:46.560> papers<00:09:46.959> that<00:09:47.040> they<00:09:47.120> might<00:09:47.360> write<00:09:47.600> so<00:09:48.160> i 00:09:48.230 --> 00:09:48.240 align:start position:0% or in papers that they might write so i 00:09:48.240 --> 00:09:49.750 align:start position:0% or in papers that they might write so i think<00:09:48.399> just<00:09:48.560> to<00:09:48.720> establish<00:09:49.200> uh<00:09:49.440> more<00:09:49.600> of<00:09:49.680> a 00:09:49.750 --> 00:09:49.760 align:start position:0% think just to establish uh more of a 00:09:49.760 --> 00:09:51.910 align:start position:0% think just to establish uh more of a personal<00:09:50.240> connection<00:09:51.120> will<00:09:51.279> be<00:09:51.760> really 00:09:51.910 --> 00:09:51.920 align:start position:0% personal connection will be really 00:09:51.920 --> 00:09:53.910 align:start position:0% personal connection will be really helpful<00:09:52.240> to<00:09:52.399> everybody<00:09:52.720> here<00:09:53.120> um 00:09:53.910 --> 00:09:53.920 align:start position:0% helpful to everybody here um 00:09:53.920 --> 00:09:55.110 align:start position:0% helpful to everybody here um i<00:09:54.080> think 00:09:55.110 --> 00:09:55.120 align:start position:0% i think 00:09:55.120 --> 00:09:57.670 align:start position:0% i think there's<00:09:55.360> so<00:09:55.600> much<00:09:56.000> new 00:09:57.670 --> 00:09:57.680 align:start position:0% there's so much new 00:09:57.680 --> 00:10:01.350 align:start position:0% there's so much new there's<00:09:57.920> so<00:09:58.080> many<00:09:58.320> problems<00:09:59.200> uh<00:09:59.839> so<00:10:00.000> many 00:10:01.350 --> 00:10:01.360 align:start position:0% there's so many problems uh so many 00:10:01.360 --> 00:10:03.269 align:start position:0% there's so many problems uh so many there's<00:10:01.519> so<00:10:01.680> many<00:10:01.839> unknowns<00:10:02.560> in<00:10:02.640> this<00:10:02.880> in<00:10:03.040> this 00:10:03.269 --> 00:10:03.279 align:start position:0% there's so many unknowns in this in this 00:10:03.279 --> 00:10:05.590 align:start position:0% there's so many unknowns in this in this business<00:10:03.760> with<00:10:03.920> this<00:10:04.080> technology<00:10:04.880> um<00:10:05.279> that<00:10:05.519> i 00:10:05.590 --> 00:10:05.600 align:start position:0% business with this technology um that i 00:10:05.600 --> 00:10:07.670 align:start position:0% business with this technology um that i think<00:10:05.760> the<00:10:06.000> upsides<00:10:06.720> are<00:10:06.959> clearly<00:10:07.279> outweigh 00:10:07.670 --> 00:10:07.680 align:start position:0% think the upsides are clearly outweigh 00:10:07.680 --> 00:10:08.949 align:start position:0% think the upsides are clearly outweigh the<00:10:07.839> any<00:10:08.160> uh 00:10:08.949 --> 00:10:08.959 align:start position:0% the any uh 00:10:08.959 --> 00:10:10.710 align:start position:0% the any uh uh<00:10:09.519> any 00:10:10.710 --> 00:10:10.720 align:start position:0% uh any 00:10:10.720 --> 00:10:12.949 align:start position:0% uh any i<00:10:10.880> don't<00:10:11.279> yeah<00:10:11.600> maybe<00:10:11.839> i'll<00:10:12.000> allow<00:10:12.320> that<00:10:12.560> the 00:10:12.949 --> 00:10:12.959 align:start position:0% i don't yeah maybe i'll allow that the 00:10:12.959 --> 00:10:14.150 align:start position:0% i don't yeah maybe i'll allow that the the<00:10:13.120> part 00:10:14.150 --> 00:10:14.160 align:start position:0% the part 00:10:14.160 --> 00:10:16.230 align:start position:0% the part but<00:10:14.640> yeah<00:10:15.040> i<00:10:15.120> think<00:10:15.440> i<00:10:15.519> think 00:10:16.230 --> 00:10:16.240 align:start position:0% but yeah i think i think 00:10:16.240 --> 00:10:18.470 align:start position:0% but yeah i think i think the<00:10:16.560> the<00:10:17.120> the<00:10:17.279> upside<00:10:17.680> of<00:10:17.839> sharing 00:10:18.470 --> 00:10:18.480 align:start position:0% the the the upside of sharing 00:10:18.480 --> 00:10:21.670 align:start position:0% the the the upside of sharing information<00:10:19.200> uh<00:10:19.920> is<00:10:20.320> as<00:10:20.560> far<00:10:20.880> outweighs<00:10:21.360> any 00:10:21.670 --> 00:10:21.680 align:start position:0% information uh is as far outweighs any 00:10:21.680 --> 00:10:22.870 align:start position:0% information uh is as far outweighs any uh 00:10:22.870 --> 00:10:22.880 align:start position:0% uh 00:10:22.880 --> 00:10:24.790 align:start position:0% uh any<00:10:23.120> concern<00:10:23.519> about<00:10:23.920> uh 00:10:24.790 --> 00:10:24.800 align:start position:0% any concern about uh 00:10:24.800 --> 00:10:26.630 align:start position:0% any concern about uh you<00:10:24.880> know<00:10:25.360> ip<00:10:25.839> or<00:10:25.920> anything<00:10:26.160> like<00:10:26.320> that<00:10:26.480> i 00:10:26.630 --> 00:10:26.640 align:start position:0% you know ip or anything like that i 00:10:26.640 --> 00:10:28.310 align:start position:0% you know ip or anything like that i think<00:10:26.800> i<00:10:26.880> think<00:10:27.040> we<00:10:27.200> can<00:10:27.519> we<00:10:27.680> can<00:10:28.079> you<00:10:28.160> know 00:10:28.310 --> 00:10:28.320 align:start position:0% think i think we can we can you know 00:10:28.320 --> 00:10:29.829 align:start position:0% think i think we can we can you know balance<00:10:28.720> that<00:10:29.040> and 00:10:29.829 --> 00:10:29.839 align:start position:0% balance that and 00:10:29.839 --> 00:10:31.670 align:start position:0% balance that and teach<00:10:30.160> each<00:10:30.320> other<00:10:30.560> how<00:10:30.720> to 00:10:31.670 --> 00:10:31.680 align:start position:0% teach each other how to 00:10:31.680 --> 00:10:33.990 align:start position:0% teach each other how to use<00:10:31.839> the<00:10:32.000> tools<00:10:32.560> learn<00:10:32.720> the<00:10:32.880> algorithms<00:10:33.760> true 00:10:33.990 --> 00:10:34.000 align:start position:0% use the tools learn the algorithms true 00:10:34.000 --> 00:10:35.829 align:start position:0% use the tools learn the algorithms true develop<00:10:34.720> open<00:10:34.880> source<00:10:35.200> projects<00:10:35.519> together 00:10:35.829 --> 00:10:35.839 align:start position:0% develop open source projects together 00:10:35.839 --> 00:10:37.430 align:start position:0% develop open source projects together potentially<00:10:36.560> yeah<00:10:36.720> i<00:10:36.800> think<00:10:36.959> open<00:10:37.200> source<00:10:37.360> is 00:10:37.430 --> 00:10:37.440 align:start position:0% potentially yeah i think open source is 00:10:37.440 --> 00:10:39.829 align:start position:0% potentially yeah i think open source is the<00:10:37.519> key<00:10:37.680> work<00:10:37.839> here<00:10:38.000> because<00:10:38.320> i<00:10:38.399> think<00:10:39.040> the 00:10:39.829 --> 00:10:39.839 align:start position:0% the key work here because i think the 00:10:39.839 --> 00:10:41.430 align:start position:0% the key work here because i think the this<00:10:40.079> group<00:10:40.320> is<00:10:40.480> coming<00:10:40.720> together<00:10:41.040> around 00:10:41.430 --> 00:10:41.440 align:start position:0% this group is coming together around 00:10:41.440 --> 00:10:43.509 align:start position:0% this group is coming together around shared<00:10:41.760> open<00:10:42.000> source<00:10:42.480> and<00:10:42.640> we<00:10:42.800> have<00:10:43.040> academics 00:10:43.509 --> 00:10:43.519 align:start position:0% shared open source and we have academics 00:10:43.519 --> 00:10:45.430 align:start position:0% shared open source and we have academics producing<00:10:44.000> it<00:10:44.160> we<00:10:44.320> have<00:10:44.399> practitioners 00:10:45.430 --> 00:10:45.440 align:start position:0% producing it we have practitioners 00:10:45.440 --> 00:10:47.350 align:start position:0% producing it we have practitioners taking<00:10:45.680> it<00:10:45.839> and<00:10:45.920> hopefully<00:10:46.560> by 00:10:47.350 --> 00:10:47.360 align:start position:0% taking it and hopefully by 00:10:47.360 --> 00:10:49.829 align:start position:0% taking it and hopefully by uh<00:10:47.760> providing<00:10:48.160> feedback<00:10:48.880> yeah<00:10:49.200> uh<00:10:49.519> we'll 00:10:49.829 --> 00:10:49.839 align:start position:0% uh providing feedback yeah uh we'll 00:10:49.839 --> 00:10:52.550 align:start position:0% uh providing feedback yeah uh we'll improve<00:10:50.399> uh<00:10:50.640> our<00:10:50.959> common<00:10:51.440> open<00:10:51.680> source<00:10:52.000> and 00:10:52.550 --> 00:10:52.560 align:start position:0% improve uh our common open source and 00:10:52.560 --> 00:10:54.870 align:start position:0% improve uh our common open source and share<00:10:52.720> as<00:10:52.880> much<00:10:53.120> as<00:10:53.200> we<00:10:53.360> can<00:10:53.680> absolutely<00:10:54.399> yeah 00:10:54.870 --> 00:10:54.880 align:start position:0% share as much as we can absolutely yeah 00:10:54.880 --> 00:10:56.630 align:start position:0% share as much as we can absolutely yeah well<00:10:55.120> thanks<00:10:55.360> adam<00:10:55.760> it's<00:10:56.000> great<00:10:56.160> to<00:10:56.240> be<00:10:56.399> here 00:10:56.630 --> 00:10:56.640 align:start position:0% well thanks adam it's great to be here 00:10:56.640 --> 00:10:58.150 align:start position:0% well thanks adam it's great to be here and<00:10:56.800> i'm<00:10:56.959> looking<00:10:57.200> forward<00:10:57.519> to<00:10:57.839> to<00:10:58.000> your 00:10:58.150 --> 00:10:58.160 align:start position:0% and i'm looking forward to to your 00:10:58.160 --> 00:10:59.750 align:start position:0% and i'm looking forward to to your presentation<00:10:58.720> all<00:10:58.880> right<00:10:58.959> thank<00:10:59.120> you<00:10:59.440> yeah 00:10:59.750 --> 00:10:59.760 align:start position:0% presentation all right thank you yeah 00:10:59.760 --> 00:11:17.750 align:start position:0% presentation all right thank you yeah we're<00:11:00.079> happy<00:11:00.240> to<00:11:00.399> host<00:11:00.959> thanks 00:11:17.750 --> 00:11:17.760 align:start position:0% 00:11:17.760 --> 00:11:19.839 align:start position:0% you