ja3k

Location: Seattle

Remote: Preferred

Willing to relocate: No

Technologies: llm apis, postgres, python, typescript, react, materialui, fastapi, aws (s3, ec2, beanstalk, lambda, cloudfront)

Résumé/CV: https://ja3k.com/resume.pdf

Email: [email protected]

I've been working on llm codegen the last 2 years at mentat.ai and would like to keep working on it.

Sorry about that. We updated the blog with attribution and put an attributing comment in our code base where we use your benchmarking prompts. We'll probably delete our implementation of your response format later today since we just had it for benchmarking.

A tool to cross poast to different social networks from the command line.

Location: Seattle Remote: remote or in person in the Seattle area Willing to relocate: no Technologies: Python, Javascript, Typescript, Clojure, Postgres, React, Express, Tailwind, AWS, gcloud, Vimscript, ChatGPT Resume: ja3k.com/resume

I've made one GPT wrapper: https://github.com/jakethekoenig/ChatVim and am interested in making them professionally. Not sure whether I'm early or late.

Location: Seattle Remote: Open to in person in the Seattle area or remote. Willing to relocate: no Technologies: Python, Typescript, Javascript, Clojure, Vimscript, chatGPT, copilot, git, vim, tmux, aws, gcloud, React, Flask, Numpy, Express, Terraform, Postgres Résumé/CV: ja3k.com/cv Email: [email protected]

Location: Seattle, WA Remote: preferred Willing to relocate: No Technologies: NodeJs, Typescript, Python, Clojure, AWS, gcp, Ubuntu Resume: https://ja3k.com/cv email: See resume

I've become sort of a generalist and am open to hearing about any opportunities but some things I'd be particularly suited for or interested in:

Roles where I can apply my mathematical experience. I've written two papers in random matrix theory. I don't expect most employers will be interested in the results themselves but I have strong linear algebra and probability fundamentals.

Full stack software engineer roles.

I'm also interested in trading. As a side project I traded just over a million dollars of volume last month and made a $400 profit.

I see a lot of posts https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... about alcohol and in particular the WHF policy brief was very highly up voted with a title echoing a misrepresentation they make.

I wrote a children's book: https://www.amazon.com/Exponential-Growth-Babies-Jake-Koenig... The subject matter may appeal to any parents here.

For a bit I had google ads on my blog: https://ja3k.com but I only made $15 in a couple months so now I just have an ad for my book on my blog posts.

To be honest I hated the nags. Little pop ups on amazon.com and occasional e-mails that didn't even have unsubscribe links. If you want to donate to charity great, but as your customer I don't want to hear about it.

I think the Canadian report it references is here: https://www.ccsa.ca/sites/default/files/2023-01/LRDG%20Lifet...

I might be being dumb but does anyone understand the graphs on page 18? There are multiple lines for different consumption levels but the x axis is also consumption level? I also don't understand why everything is piece-wise linear. It makes me think they don't actually have very high quality data? Maybe people just self selecting buckets?

Not that related to tech but I remember people were interested in an earlier post about problems with infinite utilitarianism.

In my data my REM and deep sleep percentages both improved while my total time asleep went down (in my mind a win-win). Subjectively my sleep velocity also improved.

I don't have data for what happens when I have 2+ drinks but that definitely has a deleterious effect on my sleep. Especially after 3.

About a year ago I started drinking a glass of wine every night because of a self experiment I did suggesting it was good for my sleep (https://ja3k.com/blog/wine). The title concerned me but if you look at the graphs there's basically no difference between 0 and 1 for either gray or white matter.

A note on photos: If you use Google Photos you can have photos with a specific person automatically added to a shared album. So my wife can always see every photo I take of her or our daughter. I just did it for day to day convenience but it'll also come in handy if I pass since she'll have all the most important photos already.

Though after seeing this thread I went ahead and just gave her access to my photos once I go inactive using this service: https://support.google.com/accounts/answer/3036546?hl=en

I've found the phone's small keyboard makes a lot of the silly vim shortcuts I've learned over time actually worth it.

I've been programming on my phone with termux + vim. It's been sort of a delight.

> Over time, this process will starve these com­mu­ni­ties. User atten­tion and engage­ment will be shifted into the walled gar­den of Copi­lot and away from the open-source projects them­selves

The author seems to be implying that since Copilot can reproduce the code of open source repository X in certain scenarios there'd be no reason for programmers to learn/use/engage with repository X. But this is silly. Maybe some open source repositories could be tab completed with a little prompting but people will presumably choose to add a dependency instead of tab completing the code of express or something.

I got hired for my current role after posting on "Who wants to be hired". Around ~10 companies reached out.

I couldn't convince my mother in law it was more impressive than photoshop.