donretag

Location: Los Angeles

Remote: Yes

Willing to relocate: No (for now)

Technologies: Elasticsearch, OpenSearch, Vespa and practically every DB under the sun. Language agnostic but the bulk of my work has been in Java/Kotlin/Scala, Python, and some Typescript. AWS, NLP, LLM, Transformers.

Email: https://url.dev/m/xpQpdwn/

About me: experienced software engineer passionate about search technologies. Looking to break into proper machine learning (I have experience) and/or build something interesting/meaningful. Long-time backend developer, inexperienced in backend. Been navigating the startup space lately, but it is time to give up and focus on finding something permanent.

This behavior is pretty much the state of the American two party political system. For example, Democrats had countless attempts to make abortion a constitutional right, but if they did, they could no longer count on using that subject for fundraising against the "enemy". Using Democrats as an example, both parties are guilty.

Vector search is not exclusively in the domain of text search. There is always image/video search.

But pre-filtering is important, since you want to reduce the set of items to be matched on and it feels like Elasticsearch/OpenSearch are fairing better in this regard. Mixed scoring derived from both both sparse and dense calculations is also important, which is another strength of ES/OS.

Curious about the lack of Vespa, especially given the thoroughness of the article and its long-time reputation. OpenSearch is also missing, but perhaps it can be considered being lumped in with Elasticsearch due to them both being based on Lucene. The products are starting to diverge, so would be nice to see, especially since it is open-source.

For the performance-based columns, would be also helpful to see which versions were tested. There is so much attention lately for vector databases, that they all are making great strides forward. The Lucene updates are notable.

Great article. Well done. Have you considered adding an RSS feed for your content?

I spent 3 months in South Africa, primarily in Cape Town. Visted Mabu Vinyl, but sadly did not purchase anything (was backpacking, so traveling light). Owner was lovely.

The documentary is not exaggerating the fact that Rodriguez was immensely popular in South Africa. Heard his music playing on iPods in bars, young 20 somethings that not only listened to him, but so did their parents.

The movie does exaggerate his disappearance. Rodriguez toured Australia, so he knew he had some popularity.

A few years later, I met the owner of Light in the Attic records, the reissue label that repressed Rodriguez's albums. They acquired the rights before the documentary, and even they were amazed at the popularity of his music after the movie. He told basically that Rodriguez paid for his house due to the sales.

That is what I assumed as well, until one day I got hit by a bug involving Extension Mechanisms for DNS (EDNS). Never knew it existed. All of a sudden DNS was failing and could not understand why. Took me a long time to fix the issue.

Not too far from the Remote Lounge was a Clockwork Orange themed bar named Korova Milk Bar around the same time.

Since Open Source has been established in the tech ethos for a while now, any deviation has been met with derision. It seems like the community has been more tolerant of these "open" licenses as of late. While must of the hate for projects that do not fit the FOSS standard is mostly unwarranted, hopefully we are not moving quickly in the "open" direction.

Here is another article on LLaMa2: https://opensourceconnections.com/blog/2023/07/19/is-llama-2...

Very true. Same height/length, and wider!

https://www.gsmarena.com/compare.php3?idPhone1=12380&idPhone...

Miss my Sony Xperia Compact. Would love to see that form factor again.

If anyone from posthog is reading this, please fix your RSS feed. The link actually points back to the blog homepage.

The overwhelming majority of users on most social networking sites are lurkers. Those lurkers did not vote. I would assuming the majority of users do not even know about the changes or the blackout.

Title comes from the reporting on Deadline, an industry site https://deadline.com/2023/05/wga-strike-scab-reporting-web-s...

That was pretty much my experience. Immigrant family and since I went to high school in Europe, I did not have much of a choice for university and ended up at CUNY. Worked evening in supermarkets for the first couple of years, then I was able to get an internship due to my grades, but basically installing computers, not development. There were not many internships available. Eventually went to a top engineering uni for grad school. Despite all of this, I am still considered privileged and white. When you have no circle, no network, no money, when no one in your family went to college, everything is more difficult.

I assumed it had something to do with the color and tried to find patterns. Turns out you need knowledge of the London Metro system. It would be nice if the title reflected that.

My personal definition of Big Data has always been when you gather/store data without having a planned use for it. Do we need this data? Don't know, let's just store it for now.

The article does allude to this definition when it states that "Most data is rarely queried". We have become data hoarders. Technology has made it easy (and relatively cheap) to store data, but the ideas of what to do with this data have not scaled in comparison.

"Warrior needs food badly"

"Wizard is about to die"

There is a Slack group related to search at http://www.opensourceconnections.com/slack

Great community. I would love to join something from the start, but now is not the right time.

There is some gamification like this at https://wandrer.earth/

I've been doing the same using Strava Heatmaps, cycling the Hollywood Hills. Lots of very steep dead-end streets. Cannot just wing it since you never know when there might be a street coming up, new streets need to be planned.

Links to actual content simply go back to https://angel.co/jobs. Wellfound is basically just a landing page at this point.

When searching for a job, especially in a startup, I found Angel List worthless. All primary results were basically massive companies like Amazon.

I worked at Ticketmaster when covid first hit. Since the beginning, management has always been positive and re-iterating that the cash reserves are substantial and the company can endure the loss of revenue. Then the first layoff hit. Same re-iteration. We are good. Then came the second layoff.

Yet both Netflix and Disney+ are introducing cheaper ad-supported tiers. Consumers want cheaper services, at the cost of advertising.

Hacker News lives in a bubble of users that understand and value privacy, knows about browsers such as Brave, etc. The rest of the world just wants their influencer content.

Online applications are indeed ridiculously long, yet somehow, when a recruiter contacts you from LinkedIn or other method, they do not require any of this info. No formal resume, no references, nothing.

A large part of the problem in the us is the required "Voluntary Self-Identification" information. It is voluntary, but it still needs to be filled out, even if the response is Decline to Answer. Should be optional. Companies are now taking an extra step and asking about pronouns and other identifiers, beyond what is required by law. I just want to complete this application and move on.

I explicitly have on my profile that I am not interested in Amazon, but their recruiters still contact me. Recruiters still not read profile, it's basically spray and pray.

That said, it's been a couple of weeks since I have heard from Amazon. There truly must be a hiring freeze going on

Autocomplete requires CPU cycles for the lookup and "jump to definition" requires an index to be built somewhere (CPU again, but also mem/disk space). Modern computers can handle both efficiently nowadays, but not a simple task back then.