Skip to main content
Search engine

Search engine

Software typeWikipedia

Search complete. 43 mentions across 12 episodes found for "Search engine".

Sep 15, 2026

HackerNoon AINARRATOR
0:00
This audio is presented by Hacker Noon, where anyone can learn anything about any technology.
HackerNoon AINARRATOR
0:05
Six Lessons From Building an AI-Powered Marketplace Search Engine by Ohad Farkash.
HackerNoon AINARRATOR
0:10
Two years ago, searching a large marketplace catalog in anything other than English or Chinese was close to useless.
HackerNoon AINARRATOR
0:17
Type a plain description of what you want in Hebrew, Arabic, or Polish and you get a handful of results, most of them irrelevant.
HackerNoon AINARRATOR
0:27
The problem is not that they are missing, it is that the words a shopper uses and the words a seller writes in a listing title are different words, and the gap gets wider the further you move from English.
HackerNoon AINARRATOR
0:37
That is a translation problem in the loose sense, and an LLM is obviously good at it.
HackerNoon AINARRATOR
0:42
So I built a search engine that takes a plain language query in any of twelve languages, converts it into the terms sellers actually list under, queries the marketplace API, and ranks what comes back by real rating and order volume rather than ad spend.
HackerNoon AINARRATOR
0:56
The language part worked roughly as expected.
PJ VogtHOST
49:59
Kidding, kidding.
PJ VogtHOST
50:01
However, we did put out a call to the hundreds of thousands of listeners in Search Engine nation and asked to hear from people who have visited Japan, wanting to check our work here.
PJ VogtHOST
50:12
And once we asked, we unsurprisingly got an inbox-melting number of replies, which looked like a lot to sort through.
PJ VogtHOST
50:19
But fortunately, Piper Dumont, Search Engine's chief data scientist and also intern, collated the responses.
PJ VogtHOST
50:25
It turns out that our listeners' reasons for visiting map insanely well onto what we've just learned.
PJ VogtHOST
50:32
For instance, nearly half of our respondents mentioned culture.

6 MINS LATER

PJ VogtHOST
56:57
I'm sorry we couldn't play them all.
PJ VogtHOST
56:58
We did listen to them all, and we're gonna do more of this in the future.
Evan RobertsHOST
7:53
When the game is over, go to the press box.
Evan RobertsHOST
7:55
That's where you can do your podcast.
Evan RobertsHOST
7:57
Or do a drive home.
Evan RobertsHOST
7:58
I've perfected the drive home.
Kevin SheehanHOST
28:14
There are plenty of people my age that don't like Commanders either." But he's w- he writes, "I'm a Washington fan living in Ohio.
Kevin SheehanHOST
28:20
I found you and Tom," he listens to the podcast as well, "in the last year and really love your show.
Kevin SheehanHOST
28:27
It makes sense to me that fans are swayed by what they've seen recently, but how do people who cover the sport, who ranked Washington in the top 10 last year in preseason rankings now have them in the low to mid-20s when the only reason last season went the way it did was because of injuries? Make it make sense to me." Um, this is kind of a conversation on repeat to a certain degree.
Kevin SheehanHOST
28:56
Uh, first of all, I, I, I know we talk power rankings, and by the way, Max and I will do NFL power rankings for the first time next Thursday, not today.

1 HR 6 MINS LATER

Kevin SheehanHOST
95:08
We can come back to some of this next week.
Kevin SheehanHOST
95:10
Let's talk some football with Charge.
Kevin SheehanHOST
95:12
By the way, Charge's new podcast is Fantasy Football World.
Kevin SheehanHOST
95:17
You can go to ffworldshow.com, uh, to link to it.
Connor ShortenHOST
36:28
Yeah, so I guess, so I don't mean to change the topic too much, but something I'm really interested in, we worked on a paper called IR Papers, where it was about, you know, we curate like an index of information retrieval papers and we're doing research just on, you know, search on that.
Connor ShortenHOST
36:44
And something I haven't published, but I was comparing this with parallel web search, not to like say we've averse parallel.
Connor ShortenHOST
36:51
I love parallel.
Connor ShortenHOST
36:52
It's one of my favorite web search.
Connor ShortenHOST
36:53
That's why I was using it to compare.
Connor ShortenHOST
36:55
But, and I'm trying to answer this question of like, Is it worth it building your own local knowledge base with information that is on the web? And so I think that's an interesting thing as we're talking about like auto index and LLMs and structure index and all the like advancement you can put into your own curated indexing.
speaker_1HOST
9:18
And this is achieved using the second tool,
speaker_0HOST
9:20
the web search tool.
speaker_0HOST
9:21
Web search.
speaker_1HOST
9:22
Yeah.
speaker_1HOST
9:23
It completely bypasses the limitations of the static training data by allowing the model to retrieve fresh external information at runtime.
speaker_0HOST
9:31
OK, here's where it gets really interesting, because the source material highlights this feature called automatic query generation.
speaker_1HOST
9:37
Yes, that's a big one.
speaker_0HOST
9:39
And I want to slow down and look at how this works, because as a user, I don't have to explicitly type like, hey, AI, please initiate the web search protocol to look up the current stock price.
Xiaochang LiGUEST
39:02
So my next book that I have been working on is, you know, kind of broadly...
Xiaochang LiGUEST
39:09
A history of the search engine and trying to think about the history of information retrieval and how it was that we got to this set of technologies that we understand to be like modern web search.
Xiaochang LiGUEST
39:22
In part because I think one of the things that we tend to forget is because it's so deeply embedded, right? Like I like to think like the history of the obvious thing that we don't question is that our understanding of like, you know, as everyday people and consumers, our understanding of what the internet is comes in the shape that the search engine gave it to us.
Xiaochang LiGUEST
39:48
And so I thought it would be important to really think about like, well, then what were the decisions that push us towards creating information retrieval in this way? And how did that shape the kind of industries and platforms that we see today?
Miranda MelcherHOST
40:06
Well, that certainly sounds interesting.
Isaac RoppHOST
37:49
And we're just sitting over here.
Isaac RoppHOST
37:51
We kind of look like a podcast.
Isaac RoppHOST
37:52
We're unmarked.
Isaac RoppHOST
37:53
You can't even see our.
Isaac RoppHOST
38:33
Anyway, so we're just sitting over here, you know, just kind of shooting the breeze.
Isaac RoppHOST
38:37
And, you know, nobody knows who we are.
Isaac RoppHOST
38:39
We just look like a sports betting podcast or something.
Isaac RoppHOST
38:41
I'm in my swim trunks.
Cassie ClarkHOST
1:45
Let's start with Google.
Cassie ClarkHOST
1:46
Last week, Google announced several new personalization features across Search, Discover, and Google News.
Cassie ClarkHOST
1:53
One of the most interesting for marketers, though, is the expansion of preferred sources.
Cassie ClarkHOST
1:58
Google announced preferred sources a while ago, and it allows people to select publications they want to see more frequently in Search.
Cassie ClarkHOST
2:06
Now publishers can actually add a preferred sources button to their own websites to make it easier.
Cassie ClarkHOST
2:12
A reader can click that button, add the website as a preferred source, and then return to the publisher's website.

10 MINS LATER

Cassie ClarkHOST
11:58
It's just become the source that people and machines want to find.
Cassie ClarkHOST
12:02
And then also making sure your brand is discoverable enough whenever an audience goes looking for you.
Julian GoldieHOST
0:08
There's a brand-new update to Gemini Notebook, and it's huge.
Julian GoldieHOST
0:11
Your research can now follow you right into Google Search.
Julian GoldieHOST
0:14
This is new, this is big, and it's stupidly easy to use.
Julian GoldieHOST
0:17
Stick around because SEO is about to get a lot faster for you.
Julian GoldieHOST
1:35
So if you're doing SEO research, you could drop in ten competitor articles and ask Notebook to turn that into a report showing content gaps you can target for new keywords.
Julian GoldieHOST
1:44
Update two, and this is the one that really matters for SEO.
Julian GoldieHOST
1:47
Notebooks now show up inside AI mode in Google Search.
Julian GoldieHOST
1:51
Let that sink in for a second.

2 more episodes mention Search engine.

Create an account to see the whole feed, search across every transcript, and follow the entities you care about.

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.