pip (@smallboxswe) header
pip (@smallboxswe) profile picture

I launder bits to make money | for the record, I work on query engines

venv
Joined June 2022
Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)Photo by pip (@smallboxswe)
pip retweeted
Meet MariaDB Enterprise Platform 2027 — built for the agentic era, with a new look to match ✨ Agents are arriving faster than most data stacks are ready for, and every enterprise putting them to work is adding traffic its data layer was never built to carry. MariaDB Enterprise Platform 2027 is purpose-built for that load, and for the applications your teams already run — so you can build agents faster, run them faster, work from one live data layer, scale without the rebuild, and maintain security and control across every workload. And because the platform has evolved far beyond a traditional database, the brand has evolved with it: open by design, unified across every workload, and built for the agentic enterprise. We’re the same team you depend on, with a new look and feel as bold as what we’re building for you. MariaDB is the Unified Open Database Platform for the Agentic Enterprise. Explore it here: mariadb.com/newsroom/press-r…
3
6
2
23
2,355
I know this makes me a sicko but I love C. I love reading C. I love writing C. I love everything about C. I should not be ALLOWED to write C, I get it, but still.
24
19
4
308
16,115
nodthini
is there a word between yes and no
39
393
12
3,480
49,348
ಕನ್ನಡ ರಾಜ್ಯೋತ್ಸವದ ಶುಭಾಶಯಗಳು! Jaya Karnataka Mathe! 😂🙏🏼
3
90
4
553
16,131
This is Drake vs. Kendrick for people who know what a data lake is
Replying to @RamaswmySridhar
I don't think we should fight. The two companies are complementary. Since you're not doing this AI stuff, I really think Snowflake should consider getting its AI tokens through Databricks. You can get both cheap OSS models as well as frontier (OAI/Ant/Gemini/Grok) models from our platform. It all goes through Unity Gateway which has Smart Routing, Budgets, and Access Control. You can even use Omnigent with any harness, including CoCo. Most of our customers used this to reduce token spend ~30% last few months while token count went up. Genie lets you then track, control, manage that spend, all your financials, marketing, etc. Quick POC?
15
38
4
979
78,352
pip retweeted
if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1 now you have evil qwen recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input
515
1,325
779
22,406
4,536,096
pip retweeted
cognition's ceo is my benchmark for being qualified as a smart person
70
137
73
4,484
679,347
I, for one, consider 4 a perfectly fine integer between 1 and 39402006196394479212279040100143613805079739270465446667946905279627659399113263569398956308152294913554433653942642
58
1,169
35
29,033
331,224
Of course, that’s your contention. You’re a first-year machine-learning engineer. You just got finished readin' Attention Is All You Need, probably watched a Karpathy video too. So now you’re convinced everything is just transformers and scaling laws. That's gonna last until next month when you discover convolution, and then you're gonna be talkin' about inductive biases and locality and how CNNs were actually incredibly compute-efficient for vision. Then you're gonna read the FlashAttention paper, and suddenly everything's about IO complexity and SRAM and how FLOPs don't matter because the whole goddamn thing is memory-bandwidth bound. That'll last until somebody shows you an MoE model, and then you're gonna be in here regurgitating DeepSeek, talkin' about expert parallelism and active parameters and how dense models are economically obsolete. Then six months from now you'll write one shitty Triton kernel, look at an Nsight trace for the first time, and start telling everybody Python isn't actually the bottleneck because the GPU is asynchronous. And by next year you're gonna be standing right here explaining to me that your 400-billion-parameter model is fast because you turned on CUDA graphs.
Ben Affleck reveals he writes Python, understands convolutional neural networks, worked extensively with GPUs, and used his celebrity status to get private looks at Google and OpenAI’s video models “I’ve always been kind of into computers since I was young. Then, when film started to move from analog film to digital, I became more interested in that aspect of it. The visual-effects workflow for many years has included machine learning, so I can write pretty shitty Python scripts and stuff like that. “With convolutional neural networks, which were the precursors to what the transformer can do, which is much more computation simultaneously, you would do things like look at what’s called a tensor. That’s the numerical translation of a visual image in numbers, like the batch number, the frame number and the red, green and blue values of each pixel in each frame. It’s just that simple. That numeric is called a tensor. “You’d use a convolutional neural network to identify patterns that reveal what’s called edge detection or feature extraction, which is identifying patterns well enough to know, this is where the window ledge is, so we can more easily take the green-screen image out and replace it with something. “That was familiar to me early on because, prior to Artists Equity, I had a small visual-effects company. I’ve worked with GPUs a lot too. The visual-effects guys said, ‘Hey, you should see. There are a couple: Google and this other company, OpenAI, are doing really interesting stuff with transformers in video.’ “I’ve learned that I can actually just call up and go, ‘Hey, it’s Ben Affleck. Can I come see what you’re doing?’ Sometimes people say yes, to my astonishment.”
274
1,061
139
12,102
550,058
Ramanujan wasn’t skipping proofs. He was hiding his chain of thought to prevent distillation. Hardy: “Show your work.” Ramanujan: “Sorry, reasoning tokens aren’t exposed via the API.”
80
710
70
9,812
997,478
Whatever the situation, thought of this meme always brings a smile to my face.
38
386
20
6,332
134,730
India needs Alex Karps not Nikhil Kamaths
24
123
9
1,476
40,761
Throwback to NNT chacha banger on social science
Not just sociolists. Because it, as it's currently measured, is. NNT isn't a 'yeh sala humanities log' type person. He has written whole ass blogs on why IQ 'as is measured now' has horrendous statistics properties and is thus a very very bad measure.
4
88
5
689
10,076
80
3,364
139
63,396
633,695
13
348
23
2,871
197,129
NOW IS IT CPU BOUND, OR IS IT MEMORY BOUND?
71
507
35
14,175
288,897
Sadhguru being released onto stage every Mahashivratri
I’ll never look at another Vulture the same again 😂
345
2,581
156
28,247
830,763
Imagine the war room for ensuring deprovisioning @JeffDean’s access doesn’t take down all of Google
Sergey is the only one worthy of removing @JeffDean from all OWNERS files. Make it happen.
2
13
1
586
71,669
I am excited to announce that I, along with my long-time colleagues (@JeffDean, @OriolVinyalsML, @quocleix), are founding Discovery Loop (@DiscoLoopAI), a public benefit corporation. We aim to automate the process of discovery and progress in machine learning, science, and engineering. The four of us have helped build some of the world’s most used products and infrastructure. We’re now excited to turn our attention to this ambitious endeavor. ♾
191
288
62
6,222
706,693