Make me a X-mas Song Lyrics!

Grok:

"Oh! Rocket bells, Elon's bells,

Launching all the way,

Oh, what fun it is to fly,

In a SpaceX sleigh, hey!

A day or two ago,

I thought I’d take a flight,

And soon, Miss Doge was there,

With Dogecoin shining bright.

The rocket was sleek and fast,

Misfortune not its lot,

It shot into the sky so high,

And to Mars we got shot!

Hey, rocket bells, Elon's bells,

Launching all the way,

Oh, what fun it is to fly,

In a SpaceX sleigh, hey!

Then Donald with his MAGA hat,

Said, "Make Earth Great Again!"

He hopped aboard the sleigh,

With a plan that was quite insane.

He wanted to save the planet,

With a tweet and a shout,

But all he did was tweet,

And then his sleigh conked out.

MAGA bells, MAGA bells,

Tweeting all the way,

Oh, what fun it is to ride,

In a one-horse MAGA sleigh, hey!

A day or two ago,

The story I must tell,

I went out on the snow,

And into a snowbank fell.

Elon passed by in his sleigh,

Laughed at my sorry plight,

But then he gave a ride,

Saying, "Let's save the planet right!"

Save the Earth, save the Earth,

Saving all the way,

Oh, what fun it is to ride,

In a green open sleigh, hey!

Now the ground is white,

Go green while you're young,

Take the planet tonight,

And sing this saving song.

Just get a Tesla sleigh,

So silent, fast, and clean,

Hitch it up with solar power,

And lead the eco-dream.

Rocket bells, rocket bells,

Launching on this day,

Oh, what joy to save the world,

In a SpaceX green sleigh, hey!"

Source: x.com/nfkmobile/status/1863114691611943171

Source: x.com/nfkmobile/status/1863121012553924847

This post is bait disregard it, but in which use case do one use Haskell?

>be me

>coded OS in rust

>coded large web crawlers in GO

>hundred million python projects

>Julia AI thing

>Swift webserver

>C for some games

What do you use Haskell for?

Source: x.com/BLUECOW009/status/1863118770182451462

I have to really focus so I can see a model come out of this

Source: x.com/BLUECOW009/status/1863124647086014948

*sigh* looks like I'm making my own Vitest-based evals framework

Source: x.com/mattpocockuk/status/1862502667706093936

There's a massive opportunity for subject matter experts to create better-than-the-average-labeller LLM's for specific tasks.

Just give it good data, via RAG (cheap) or fine-tuning (expensive).

Source: x.com/mattpocockuk/status/1862578094374011094

What's the best npm package for making a pretty ascii table in the terminal?

Ideally would be responsive to terminal changing size.

Source: x.com/mattpocockuk/status/1862854416950288731

Final Words

⬙ Choose one database (e.g., MySQL).

⬘ Spend at least 1 to 2 hrs/day learning and practicing.

⬙ Overall it should take 6 to 9 months.

⬘ I suggest to also learn a programming language in addition to SQL if you want to be an application developer.

Source: x.com/swapnakpanda/status/1863082441415536779

➓ Advanced Concepts

Effort: 1 to 2 hours per day Study + Practice

Time: 4 to 6 weeks

❯ Binding variables

❯ Temporary tables

❯ Common Table Expressions (CTE)

❯ Regular Expressions

❯ Pivot data

❯ Import data

❯ External tables

Source: x.com/swapnakpanda/status/1863082439112900712

I have been working in SQL since 2004.

If I have to start learning it in 2025, I'll do this:

Source: x.com/swapnakpanda/status/1863082409060602161

15 FREE Mathematics Books

❯ Computer Science

❯ Data Science

❯ Machine Learning

1. Mathematics for Computer Science (MIT Press)

https://courses.csail.mit.edu/6.042/spring18/mcs.pdf…

Source: x.com/swapnakpanda/status/1862849502987587846

Top 150+ algorithms for the interviews

❍ C / C++

❍ Java

❍ Python

❍ JavaScript

❍ Golang

Or, any language of your choice

Source: x.com/swapnakpanda/status/1861414926373290290

Scrum's "definition of done" is one of the least agile parts of Scrum. You're done when you release. You are doing that every couple of days, aren't you? Of course, the resulting feedback will require adjustments in whatever you thought was done. Attempts to create "done" criteria up front is just waterfall thinking. You won't know if it's right until you get it into your user's/customers' hands.

Source: x.com/allenholub/status/1863083710511218837

This is a limitation of the browser. You can add a bunch of JS to make your own select, or just leave it. Customizable select will make styling these things so easy

Source: x.com/wesbos/status/1861532165701251302

People have too inflated sense of what it means to "ask an AI" about something. The AI are language models trained basically by imitation on data from human labelers. Instead of the mysticism of "asking an AI", think of it more as "asking the average data labeler" on the internet.

Few caveats apply because e.g. in many domains (e.g. code, math, creative writing) the companies hire skilled data labelers (so think of it as asking them instead), and this is not 100% true when reinforcement learning is involved, though I have an earlier rant on how RLHF is just barely RL, and "actual RL" is still too early and/or constrained to domains that offer easy reward functions (math etc.).

But roughly speaking (and today), you're not asking some magical AI. You're asking a human data labeler. Whose average essence was lossily distilled into statistical token tumblers that are LLMs. This can still be super useful ofc ourse. Post triggered by someone suggesting we ask an AI how to run the government etc. TLDR you're not asking an AI, you're asking some mashup spirit of its average data labeler.

Source: x.com/karpathy/status/1862565643436138619

Never give up !

Source: x.com/nfkmobile/status/1863071008321654967

...When Tesla gets Grok improvements FSD 42.0.69

Source: x.com/nfkmobile/status/1862958001759367516

May The Doge Be With YOU!

https://video.nostr.build/a4d7cc04d8dc6b6ffb47578cc61310ae9191677ceb2aafa74704876f98b6e51d.mp4

Source: x.com/nfkmobile/status/1862772047157362881

“Elon Musk and Donald Trump are like best friends." -Maye Musk (hope they like memes also)

Source: x.com/nfkmobile/status/1861682905753305273

added activity_constraints.py for easy hard-coded rate limiting activities (eg. must be X hours, min/max per day), as well as 24-hour activity dashboard to monitor if this is working

helps manage cost before we let it run 24/7*

*he might nap a lot at first, before we get a better sense of cost/activity, etc.

Source: x.com/yoheinakajima/status/1863069727813202398