katie
Katie

ML & Data Engineering Consultant | Reader | Knitter

I write code for https://nostr.wine

Public Key

npub1qlkwmzmrhzpuak7c2g9akvcrh7wzkd7zc7fpefw9najwpau662nqealf5y

NIP-05 Address

Profile Code

nprofile1qqsq0m8d3d3m3q7wm0v9yz7mxvpml8ptxlpv0ysu5hze7e8q77dd9fspzpmhxue69uhkummnw3ezuamfdejsz9mhwden5te0vdex2ct5wghxummnw3ezuamfdejslakzca

Publishing to

Author Public Key

npub1qlkwmzmrhzpuak7c2g9akvcrh7wzkd7zc7fpefw9najwpau662nqealf5y

Show more details

Last Notes

2024-12-16 20:38:10 GMT

Work from home life 😻 https://image.nostr.build/dcca75359a35faf7205acd72f158ef117924b1e0ed58caeb8464afec173c967d.jpg

2024-11-03 00:37:34 GMT

Still thinking about this view from last weekend - the whole park was stunning https://image.nostr.build/bc33a7c8008d5a9868f6fa6b82364450827be0b449ec912d381a378b4162980d.jpg

2024-10-30 23:48:54 GMT

I typically only get the desire to chop all of my hair off during the hotter months, but I gotta say I’m digging have short hair with sweater weather ☺️🧣 https://image.nostr.build/a1ee1786241ccebb9e321ae91ec1a542b40612e2807397b89eed975c5c6e6a87.jpg

2024-09-28 14:06:37 GMT

😻💤 https://image.nostr.build/6c6244ce03d6e847ee875deacb169ff23df006e928485a05f39b9bb3e30cfa11.jpg

2024-09-27 19:11:37 GMT

- reply

Yeah! I think the Zevin one got a lot of attention when it came out, and it seems like people generally like it. Let me know if you end up reading the Hart one and if you like it!

2024-09-27 19:10:31 GMT

- reply

They probably are distinct enough to read in parallel! Good recommendation 🙃 I’ll consider it. I’m kind of wondering if either one will hit that sweet spot for a weekend binge though

2024-09-27 18:20:30 GMT

- reply

It is a very pretty cover!

2024-09-27 18:06:23 GMT

Trying to decide on my next read, and I think I’ve narrowed it down to two. Anyone have an opinion of either that could tip the scales?
#bookstr https://image.nostr.build/3e511919ce30c6c1698908a5d8295c29cb1f6249cace4767548f6ca9ce2370bd.jpg

2024-09-27 17:23:06 GMT

All that I want to do when it’s a rainy/stormy day… well, this and reading a good book 📚 https://image.nostr.build/f5d1fdfbdfff7b7adf1d8c1e84f891580d820a1665fe6e9ddceff60603a27b03.jpg

2024-09-27 17:17:48 GMT

- reply

I have a hard rule against Friday deployments 😆 but the temptation is always high. Good luck!

2024-09-05 23:08:07 GMT

- reply

It’s beautiful!

2024-09-05 18:12:49 GMT

I love my little indoor jungle 🌱🍃🌿🪴 https://image.nostr.build/12fc4b1be5804d25108c50cbff2ad461cdb2807f48ed074b537884d81b264a82.jpg

2024-09-05 12:07:01 GMT

- reply

Yeah, I think that makes sense and I had a lot of similar thoughts when deciding if I wanted to pursue a diagnosis. Ultimately, for me the testing process was super informative, and seeing the strengths and weaknesses laid out so clearly was actually really validating. I think it’s helped me give myself more grace in those areas. I’m technically 2e, so I’m kind of like advanced software running on a sever with insufficient cpu and ram - and, to push the metaphor to its limits, I’ve been overclocking to keep up with my interests for a long time. It’s frustrating when the hard stuff is easy and the easy stuff is hard and you have no idea why. It’s hard not to just expect more of yourself and constantly let yourself down.

2024-09-05 03:02:38 GMT

- reply

I’m actually more curious about the other side of that coin. I’ve had my shit together my whole life - no one on the outside looking in would have ever thought there was a problem - but there was, and eventually there’s a wall after decades of brute forcing my way through it. And it’s really not even a question about medication, but one about self acceptance and that process.

2024-09-05 01:29:07 GMT

- reply

Thank you!

2024-09-05 01:11:58 GMT

Any late diagnosed ADHD people around? Curious what your experiences were/are with unmasking, if anyone is willing to share

2024-09-05 00:56:54 GMT

- reply

I do! I keep a lot of herbs around for various things, and I like to switch up my teas a good bit. I’m typically just mixing on demand for each brew and throwing in whatever I want that day

2024-09-05 00:53:38 GMT

- reply

I always forget about neti pots! Thanks for the reminder :) I’ll add that to the mix

2024-09-05 00:39:42 GMT

- reply

Thank you! This is just a loose leaf blend I threw together to help with my sore throat: ginger root, slippery elm bark, marshmallow root, fenugreek leaves, chamomile, and lavender - plus some honey in the mug

2024-09-05 00:29:13 GMT

- reply

Thank you!

2024-09-05 00:27:51 GMT

It’s starting to feel like fall 🍁🍂🫖 it could just be the sore throat that comes with the changing seasons though 🤧 either way, I’m here for it https://image.nostr.build/4b688cd65e617e9096d73fe2388cf83ce08af3bf3eb5ef255a08bedbd09e3f81.jpg

2024-06-06 22:13:07 GMT

Whelp. That’s an extremely cheap shot, is irrelevant, and misses the entire point. Decentralization, by definition, is not left or right, and it’s ignorant and harmful to push a narrative that it is. Not to mention totally unnecessary and inappropriate to dox an engineer. My political views differ from a lot of people on nostr, but that has nothing to do with the protocol or its usefulness for society. Seems like a swing and a miss on a rare topic that both sides should have common ground on.

2024-04-14 19:47:20 GMT

Sleepy kitties 😻 https://image.nostr.build/bedf8cc228ed25d51a143a2d4c15210dcc4de23c7c448a616811267bc481285a.jpg

2024-04-11 01:02:28 GMT

We’ve been enjoying our supply of greens 😍 https://image.nostr.build/aca7e67266ba856b6304103d9b765ff2b1e8d08fe0bea452ccc9edcf640c9251.jpg

2024-03-23 19:37:57 GMT

The real thing is way better than the emoji 🌷 https://image.nostr.build/5007103196291d4e32734706a4b2cb04847a268e748f75f711a4284107f8238b.jpg

2024-03-07 03:08:19 GMT

Really enjoying the redesign! 😍 #note1566…3vv6

2024-03-02 21:31:03 GMT

- reply

Thanks for the tip! I’ll gather some and add the extra layer of insulation

2024-03-02 21:28:15 GMT

- reply

A lot of herbs around the edges: parsley, dill, sage, cilantro, and a few others… and a few larger plants like kale and persian cucumbers. Lots of lettuces as well. I always have some volunteer grape tomatoes that show up from the prior year, too - so hoping those will pop up. This area doesn’t get as much sun as I’d like, since we have some big trees in the yard, so any larger fruits and veggies are a bit tough to grow in this spot.

2024-03-02 21:24:21 GMT

- reply

Good tip! I’ll keep that and mind and keep a lookout

2024-03-02 19:36:50 GMT

Getting a jump start on my garden! It’s been warmer than usual and I’m hoping between that and the frost cloth at least some of the seeds will take🤞but admittedly, I’m being a bit overeager 🫣 https://image.nostr.build/cc7837f3ca8c241d47430290fcfb92a32eb7aebf1fdd5851461c68b6279a0271.jpg https://image.nostr.build/9e66060c03e7a809b7bbbde9c9426e6bda680622abe1bed99b9f028f3a317adb.jpg

2024-02-27 01:29:36 GMT

Love seeing how we can make it easier/more clear how to follow creatrs 🙌 #note1kmx…clxf

2024-02-15 05:01:48 GMT

- reply

Yesss! We’re so close. Seed planting time and all the bulb plants are starting to pop up.
I’ve been hibernating with the cold 🫣

2024-02-15 04:20:56 GMT

I’m so ready for spring 🌷

2024-01-25 01:24:30 GMT

Love anything that makes monitoring prod services easier! If you run a personal, public, or paid relay and want to be notified if it’s ever down, check out the latest nostr.wine product. #note1nnl…y0x0

2023-11-21 18:09:02 GMT

- reply

I’d be pretty shocked if we had inbound issues 🤔 @npub18kz…x5sz runs a pretty tight ship with channel management

2023-11-01 21:37:34 GMT

The delay of this is largely my fault 🙈 I overcommitted to other projects and dropped the ball. BUT - we really are almost done 🙃
Very excited about this one.
If you’re a content creator and interested in helping us test a beta release, reach out to either @npub18kz…x5sz or myself. #note13v0…vc2y

2023-10-20 22:42:07 GMT

Finished! https://image.nostr.build/92353f31540828587cbb5e3ddfef874d88f9248aef2d35e5cfdb5c978493797a.jpg

2023-10-02 22:28:49 GMT

Made a friend on our evening stroll https://image.nostr.build/61e7d5cf65bd59d333f8c7d87cf006203986d22ed21ea4092ac6321888b29307.jpg

2023-09-27 02:57:13 GMT

- reply

I have one, but I usually don’t use it TBH

2023-09-27 02:56:14 GMT

- reply

It is! And it really does take a lot of getting used to. I like working in chalk pastels, so I really like the blending you can do with it.

2023-09-27 02:47:33 GMT

- reply

This was actually all in procreate. A tattoo shop in town puts these events together once a month, with a model and a lot of short poses that they cycle through. I usually just bring my iPad for convenience

2023-09-27 02:15:24 GMT

- reply

Figures! Ok - some poses I sketched at a recent figure drawing evening that a local group puts together. I’m out of practice 😆 but it was very fun https://image.nostr.build/fdade3bad73f9e74a3e1562c05a5062a6382591f9c4f4bd7131a31a4192728ba.jpg

2023-09-23 03:52:06 GMT

- reply

You had me at #penisbutter 🤣
I would love to join one of these women’s nostr nests sometime, but timing never quite works out. How long do they usually last?

2023-09-20 21:30:40 GMT

- reply

Thank you! I’ve been waiting on this artist for over a year and am so happy with it.

2023-09-20 21:15:53 GMT

Session one of two on a new tattoo 😍 https://image.nostr.build/0472186bfb8b961dcda9c3e5c2ef449f899e9c5fafab8ce0d100beb1be0956be.jpg

2023-09-20 01:22:00 GMT

Nap Goals. https://image.nostr.build/68761b1048b8917183c5e7acb3e711ad3ce2b0e91956dc73a9da69b557925bf7.jpg

2023-09-12 03:09:10 GMT

And that’s a wrap on an incredibly busy summer! I’m finally back. Anyone else struggle to stay on top of all the things in the summertime? 😅

2023-05-21 00:52:44 GMT

- reply

wss:// is the equivalent of https:// but for the websocket protocol vs http

2023-05-09 04:34:43 GMT

Wake up
To your life, to the day
Get out
Take a chance, go your own way
Feel the sun
On your cheeks, in your hair
Go on
Make yourself bare
Take it off
All that you do
For everyone else
Do it for you
Let yourself be
As you were
Before someone taught you
What to prefer
Shed your skin
Your habits, your role
You've outgrown this box
It's time to free your soul

2023-05-07 22:08:41 GMT

- reply

Breaking things is truly a skill 🤣

2023-05-03 21:13:29 GMT

- reply

Couldn’t have said it better myself :) I keep welcome in my list because I occasionally turn off other relays on my global and see who has recently joined

2023-04-27 21:33:15 GMT

Excited to join the 24hr Frog Talk tomorrow 🎉🐸🍷 #note1p4f…863j

2023-04-27 20:52:12 GMT

👀 https://tinyml.substack.com/p/navigating-the-complexities-of-llm?sd=pf

2023-04-27 03:45:58 GMT

- reply

I’ve recently gotten into a regular meditation practice (I’ve been a long time occasional meditator, with no real consistency), and it makes such a difference in my day. 15min of it before my first cup of coffee and 15min of it before I go to bed.

2023-04-27 03:35:56 GMT

- reply

Another tool I 💜for kube deployment is ArgoCD

2023-04-27 03:34:56 GMT

- reply

I have to work across a number of them depending on my client, but gitlab is my preferred way of operating when I get to build their CI/CD from the ground up AND I have plans to deploy to kube.

2023-04-27 03:11:33 GMT

- reply

We use kaniko for this - to use self hosted kube for gitlab runners.

2023-04-27 01:16:58 GMT

- reply

Welcome to nostr!! 🎉

2023-04-24 02:57:42 GMT

Why I #bitcoin. Women weren’t able to have access to bank accounts in the US if they were unmarried or without their husband’s permission until 1974 (legally granted rights in 1960 and functionally able to access them without discrimination in 1974 with the Equal Credit Opportunity Act, for anyone splitting hairs). I don’t want a financial system that allows any group of people to be oppressed by withholding access.

2023-04-22 22:14:28 GMT

- reply

I really enjoy your content!

2023-04-22 19:11:37 GMT

I hope your weekend is as relaxing as theirs! https://nostr.build/i/nostr.build_9cdb9c1d74138714a1c6abd354efdac39d6e95b27a0e240523c1ddc41b9b2a92.jpeg

2023-04-22 17:27:47 GMT

- reply

ALL OF THIS 👆

2023-04-22 17:24:03 GMT

- reply

I’m clearly the minority here 🙃 I used to be a binger. It ultimately doesn’t support my current goals. I have a lot of responsibilities to juggle, and I can’t support all of them if I give into the instinct to binge 🤷🏻‍♀️ I time box my binges these days.

2023-04-22 17:11:19 GMT

- reply

I’ve definitely been there. There are times when my brain cannot even grasp for words. I’ll be thinking spoon and can only come up with fork or knife. It just doesn’t work. Language is the piece that seems to disconnect first for me. Interesting that it seems to be the same for you.
And if it keeps getting pushed further with questions or problems it literally hurts. This has gotten worse as I’ve gotten older 🫠 and I’ve also come to the belief that no problem is worth getting into that state. When I’m close, I just have to say no. And when you’re working and living with others who have different levels of responsibilities and are on different binge cycles, it’s really important to set and respect those boundaries, IMO.

2023-04-22 16:55:41 GMT

- reply

🤣 I understand it, too - I just try to stick to shorter cycles. If I’m working on something I’m like a dog with a bone… but I’ve learned that if I don’t let it go in the evenings or weekends, I burn out. Supporting multiple clients and having a wide range of responsibilities has changed this for me a lot, too. I have to practice letting go and setting boundaries or I end up not being successful at anything.

2023-04-22 04:19:53 GMT

- reply

SO much of this resonates with me. And I can say, personally, I hope I’m striving for a greater understanding. But I often think of the symbol of a unalome, and the way it represents the “path to transcendence” in Buddhism - it’s not a straight line. And neither is getting out of local minima. I hope, as humans, we collectively strive for more. But I can only control myself, and trying to judge others by where they appear to be on that journey is fruitless.

2023-04-21 19:13:48 GMT

Your right to speak doesn’t equate to an entitlement for me to see what you said, if I choose to opt out. I think this extends to likes.
I’m not sure only zaps is for me. I might feel differently if I was getting some insane amount of likes - my experience is not the same as others. However, I think that enabling users to have more control over their experience (as long as it doesn’t force those same choices on others), is ultimately good.
I have a lot more thoughts about the pros/cons of it, and can see both sides. I’ll get to weigh those arguments when deciding for myself how to set my own configurations. But ultimately, I’m pro user choice, and this feature enables more of that.

2023-04-20 03:41:07 GMT

- reply

I’m super impressed!

2023-04-20 03:35:38 GMT

- reply

I wish I could. Whenever I get a new phone or take my case off it seems so pretty. But the rate at which I have to replace cases tells me that my clumsiness makes that luxury not worth the cost 🫠

2023-04-19 23:23:54 GMT

- reply

This is an antidote and how wonderfully grounding!!! 💜 Beautiful

2023-04-19 22:02:07 GMT

- reply

Thank you! This is awesome!

2023-04-19 15:45:03 GMT

- reply

I love this setup! Are these bought or made? Would love a tutorial on that cage. I have a ton of squirrels who like to munch on my veggies

2023-04-19 15:33:56 GMT

- reply

I ❤️ the little prince! And good morning :)

2023-04-19 03:01:59 GMT

- reply

Thanks for the super interesting convo ☺️ I’m going to be thinking about a lot of your points for awhile

2023-04-19 02:27:04 GMT

- reply

I’ve really appreciated your presence on nostr! We’re lucky to have you as a member of this community ☺️ always thoughtful and patient and kind. You’d definitely make my list, too!

2023-04-19 02:11:28 GMT

- reply

Yeah, it’s super interesting about the RLHF! Overfit is a very real problem and adjusting weights on models this large can be kind of like a butterfly effect. I think there is a TON of value it its generalization. But I’m of the opinion that it can’t or maybe shouldn’t do all tasks for itself - to me it’s just not necessarily efficient, like using a hammer on a screw. Bigger doesn’t always mean better - it will start to underperform at a certain size. TBD what that is. But let it do what it does best! Language and conceptual derivations and awesome encoding, let other models do what they’re better suited at. Kind of like how our brains work… we have separate specialized areas that we delegate tasks to when necessary. We’re building awesome components, but I’d like us to acknowledge their limitations, not to discourage the work that has been done, but to figure out the next problem that needs to be solved.

2023-04-19 01:50:54 GMT

- reply

For sure - and I think that is by design… they’re going to sell it to company’s who will want to specialize and limit it to be more in line with their specific use case. Right now, it’s just there to kind of be whatever you tell it to be 😂 like a giant ad.

2023-04-19 01:33:35 GMT

- reply

Maybe. But the problem with many of those internal representations are just straight up lies. It’s not a truth telling machine. It’s about to produce compelling speech. It’s a bit different in spaces like engineering, where the content is so well curated. But just because it identifies a pattern, does not make it correct or intelligent. It would need some secondary mechanism for testing hypotheses, and at that point we’re moving out of the LLM space and into the reasoning space. Progress will come when these spaces come together.

2023-04-19 01:27:00 GMT

- reply

Yeah, I think there is a lot of growth that will be happening in the model network architecture space. Using more precise smaller models for tasks that LLMs use to do specialized things, reasoning models (lots of good research going on here), as well as RI for experiments (but interestingly enough those require physics engines that we define, so most or the time they only can solve for that vs actual physics - this in combination with robotics and more sophisticated sensors though has a lot of promise - essentially self driving cars, and there is still a long ways to go there). Collaborative models are super cool though.
Ultimately though, I’m concerned about the data creation. I don’t want there to be a misunderstanding of how important it is that we continue to produce it.

2023-04-19 01:12:01 GMT

- reply

He’s making this claim based on the model’s architecture. Not all models in the future will have this limitation, but LLMs do based on their knowledge domain limitation.

2023-04-19 01:10:14 GMT

- reply

The solution to these problems is actually smaller models solving more specific tasks. That’s his argument. The larger the model, the more abstract - its good at connecting dots, but will start to struggle with precision. He’s not saying a model can’t do this, he’s saying large LLMs do some things well and some things less well and viewing it as a one stop shop for all intelligence is not reasonable. We need RI, we need more specialized tasks, and we need a lot more research in reasoning models.

2023-04-19 01:00:58 GMT

- reply

The same limitations apply… it’s still an LLM limited to a knowledge domain and is acting as an approximation machine. We still need some pathways for generating high quality data for it to be trained off of… if we use it’s “intelligence” as an excuse to stop producing that data, it will stop learning.

2023-04-19 00:36:35 GMT

- reply

Yeah, I agree it should help keep duplicate questions off the platform. And it’s for sure in chaptGPTs best interest that humans continue to produce high quality data, ranked by quality even, for more effective training. I think the ultimate and best solution is one where both systems work together. A massive part of why chatGPT is so good is because the curation of data on stackoverflow is so good. Garbage in, garbage out, as the old adage goes.

2023-04-19 00:16:09 GMT

- reply

Yeah, the more it has access to for sure increases it’s capabilities. Like giving it access to modify your terraform to deploy infrastructure. I think it’s really compelling, but I also think there is still a long ways to go. Another ML engineer in a community that I’m in put together a great piece on the current state of AI:
https://medium.com/@khalil.alami/the-boring-state-of-artificial-intelligence-152244075d7f

2023-04-18 23:58:45 GMT

- reply

Data sharing negotiations for and by AI systems 🤣

2023-04-18 23:57:00 GMT

- reply

For sure!! I’m not saying these things aren’t possible. But I don’t think they’re solved by an LLM alone.

2023-04-18 23:56:12 GMT

- reply

Right! But it’s still relying on a human in this context. Maybe it breaks our current ecosystem by not then storing that information into a publicly accessible place like stack overflow and makes us dependent on the knowledge or hordes. But what you’re describing still required missing data to be provided to it by a human.

2023-04-18 23:54:38 GMT

- reply

The other thing to note… models that generate data and then train themselves on that data that they generated, tend to get further and further away from meaningful content. If you had a model that wrote novels and eventually it’s training only on the models it wrote. They would start to all sound the same and converge. Part of why it is so compelling is that the data it is trained off of is created by a large number of humans with different perspectives and goals. You’d have to create a lot of individual models with niche goals and perspectives to sustainably let it just create indefinitely without new human content.

2023-04-18 23:51:41 GMT

- reply

Depends on the language and/or support. Could be that someone figure it out through a conversation or a guess or checking site package source code or decompiling code. Or if it’s truly a closed source project, via a client support rep. Ultimately someone posts about it somewhere and then chaptgpt knows the new argument.

2023-04-18 23:49:30 GMT

- reply

Also - think about security bugs that are reported. It knows about them because they get reported on centralized platforms. Security pen testers find them. Right now, it’s simply a language model… it has no ability to execute pen testing. It relies on humans to produce content (release notes, open source code, quality reports or GitHub threads) to learn. It’s still very much piggy backing off of human generated data. And not to say models won’t generate that data in the future… but it will be a series of models with different tasks that communicate with each other.

2023-04-18 23:44:56 GMT

- reply

Ok, so let’s say there is an update to a package that isn’t open source. I’m using the new version and can’t figure out why it’s breaking. It has no idea that the new version changed an argument name and the one I’m using is deprecated. Maybe if the release notes are in a place it has access to and can probabilistically relate them together. But if the information isn’t online (not conclusions from data points, but actually missing data points), it can’t solve that. That’s what a lot of stack overflow questions are. That’s way so many are so old… they are asked when the data points don’t exist in an accessible place. If you’re asking a question on stack overflow for data that already exists you’re just going to be pointed to an older ticket.

2023-04-18 23:41:41 GMT

- reply

I don’t think it’s fair to say no one understands how they work. Not being able to interpret weights is not the same thing.

2023-04-18 23:39:57 GMT

- reply

To give it access to do this it would need to 1. Read all code for all projects and understand every implementation and version and/or 2. Have access to execute code to observe and trouble shoot. It’s an LLM… it’s just synthesizing data that exists, and drawing conclusions and derivations from that. Impressive? Yes. Better than search? Yes. Has access to derived thought based on things that have happens that haven’t yet been put on the internet… no, not yet.

2023-04-18 23:36:05 GMT

- reply

It can take two concepts and combine them… which is what I do when I’m searching stackoverflow. Rarely does one post solve my problem. I read multiple and I synthesize that data into a solution. To me, that’s not the same as answering net new questions for new problems. And it is for sure a “not yet” problem, but I don’t think we’re there yet.

2023-04-18 23:27:56 GMT

- reply

We’ve had a lot of discussions about this. Chaptgpt can’t really answer any engineering questions that haven’t previously been answered on stackoverflow (or other sites, but I mean, it’s pretty much it). So it may pull some traffic from the lurkers who are just trying to get their questions answered. But if a question doesn’t already have an answer, someone will have to ask it to a human in stack overflow. Essentially, chaptgpt just offers better search for it, but doesn’t actually replace the recommendations for net new problems.

2023-04-18 21:34:12 GMT

- reply

💜

2023-04-18 21:32:05 GMT

- reply

Whoa, that sounds nifty 🐈 I’ll have to look into that

2023-04-18 21:14:03 GMT

- reply

Truth.

2023-04-18 21:11:34 GMT

- reply

I’m sorry 😢 I hope it goes smoothly and recovery is swift!

2023-04-18 20:31:43 GMT

- reply

Not exactly swing, but I used to do shag dancing in college. Purely for fun, socially, and I’m not very good 🥴 but it was always so much fun!