Dmitriy Ryaboy 🇺🇸🇮🇱
@squarecog
Followers
9K
Following
6K
Media
454
Statuses
13K
VP Eng. Recovering data engineer. Co-author of @MissingReadme. Plays with swords. Helped build this stupid place, long ago.
San Francisco
Joined June 2009
Yo dawg I heard you like dags so I put a sql query dag in your dbt dag in your airflow dag.
0
0
4
Sins of the past, coming back to haunt me. Been a while since I've seen an elephant bird in the wild, thought it's extinct...
1
0
5
A lot of folks know about Chesterton's fence these days, but not enough know about Chesterton's lamp-post. A similar parable, but that one comes with lessons on importance of concise and effective communication regarding the lamp-post's utility.
0
0
1
I'm in the url parsing business now. Is it to late to convert all of the web to gRPC?
1
0
6
I am begging you, throw "Let me explain" out of your writing tool bag.
1
0
8
Do millennials get Bull Durham references, or am I throwing all this heat without a batter?
1
0
2
It's funny because it's true...
Most orgs where search is important have this dysfunctional organizational process where people are siloed into ML and search separately. Where the search infra team owns infra, and the candidate retrieval process under strict latency constraints. Then, the ML people can write
0
0
3
The best systems book of the past decade, getting updated for the new decade. Can't wait!
Big news: I'm helping with @martinkl with a second edition of Designing Data-Intensive Applications! An early release of the first 3 chapters is now available (O'Reilly Learning subscribers only at this point) and we're hoping to finish it next year. https://t.co/SpDUBCLdLi
0
1
5
Did @nealstephenson CLANG ever share any of their code or the technical innovations? Would be cool to use an accurate swordfighting simulation engine for a variety of applications...
1
0
1
A veteran of early Twitter's fail whale wars, @squarecog joins the show to chat about the time when 70% of the Hadoop cluster got accidentally deleted, the financial reality of writing a book, and how to navigate acquisitions. check it out :D https://t.co/bXvFsgLX4x
0
2
9
Really hoping Tesla figures out self-driving soon cause Tesla drivers have pulled more inane near accidents in front of me lately than all others combined
2
0
6
Ok this is pretty slick. The api looks pretty intuitive, and everything gets transparently backed into SQLite for reproducibility. Easy control over (local) parallelism, etc. And Pydantic!
🔗 DataChain open-source release 🤖 AI-Driven Data Curation: Local models, LLM APIs 🚀 GenAI Dataset scale: Millions and billions of files 🐍 Python-friendly: Python objects instead of JSON Try it out https://t.co/CjY1NTDxRD 👇1/7
1
2
10
I'm reading "Sweating Bullets", Robert Gaskins' memoir about the creation of PowerPoint, and it's an absolute page turner. Edge of the seat stuff. This should be a classic of the "startup histories" genre. Go spend the $3.
amazon.com
PowerPoint was the first presentation software designed for Macintosh and Windows, received the first venture capital investment ever made by Apple, and then became the first significant acquisition...
0
2
4
Opt+K, "solution for day 1 of advent of code 23" and Cody will just write it. And tests, if you ask for them, too.
0
1
2
This was a fun and fairly wide-ranging conversation. (I don't actually think LLMs for languages are boring.. though I do think the fact that we can use LLMs for proteins is even more mind-boggling).
From building the data platform and Parquet at Twitter to using AI to make biology easier to engineer at Ginkgo Bioworks, @squarecog joins the show to chat about the early days of big data, the conversation that made him jump into SynBio, LLMs for proteins and more.
0
2
11
Would you say such a suggestion is "mildly" distracting, or "wildly"? So meta.
0
0
5
Once more, with feeling.
We will see that a lot of weird behaviors and problems of LLMs actually trace back to tokenization. We'll go through a number of these issues, discuss why tokenization is at fault, and why someone out there ideally finds a way to delete this stage entirely.
2
0
4
TBH my first reaction to Gemma was that it should stop lecturing me but then again... the others totally failed, it's right to tell me what's beyond its abilities.
1
0
3