Lightnews — Scholar-powered news

Reposted

Nikkolas Smith @4nikkolas.bsky.social · 1d

after millions of views and shares of my Portland Frog art. (thank you all🙏🏾) I got requests to highlight priests, and chickens, and Chicagoans, and T-Rexes, and more… all of us who refuse to bend the knee. so this is for US.
𝚂𝚝𝚊𝚗𝚍 𝚃𝚊𝚕𝚕.
𝚆𝚎 𝚆𝚒𝚕𝚕 𝚆𝚒𝚗.

260 7.2K 21K

pamelafox.bsky.social @pamelafox.bsky.social · 18m

Beautiful video about a trans activist and what Zohran hopes to enact in NYC for a more LGBTQ+ inclusive city:
www.youtube.com/watch?v=bEvV...

Until It's Done: Sylvia Rivera

YouTube video by Zohran Mamdani for NYC

www.youtube.com

pamelafox.bsky.social @pamelafox.bsky.social · 38m

I'll try some cheap models though (based off GitHub Copilot pricing, as that's what I'm using), that's a good idea. Right now gpt-5-mini, gpt-4o, and gpt-4.1 all have 0x multipliers, so they're the cheapest.

pamelafox.bsky.social @pamelafox.bsky.social · 40m

I had on my TODO list to try a local model, but I only have a 16GB RAM M1, and have had a hard time finding a local model with good enough tool calling support that'll run on my Mac. Are there any that you use? Typically has to be < 10 GB params.

pamelafox.bsky.social @pamelafox.bsky.social · 23h

I'm having a "test-off" where I ask different models to write tests for an API, then compare their failures and coverage.
So far:
* GPT-5 has 100% passing tests, 49% coverage.
* Sonnet 4.5 has 95% passing tests, 56% coverage.

What other models should I try?

1 2

pamelafox.bsky.social @pamelafox.bsky.social · 1d

I love the iNaturalist data!
I made a FastAPI app today that outputs "phenology" charts so you can see the active months for bees in the bay area.
(Most bees peak around May, like the Pyrobombus bumble bees)

3

pamelafox.bsky.social @pamelafox.bsky.social · 1d

For the CS exams at Berkeley in 2021, students hand-wrote their answers, and GradeScope OCR'd them as best it could. We still had to hand-grade many of the digitized answers.. I once graded 500 regular expression answers in one sitting. Never doing that again!

5

pamelafox.bsky.social @pamelafox.bsky.social · 1d

I'm trying Postgres with PostGIS in Python today. Should I use GeoAlchemy, or can I just use SQLAlchemy2 on its own?

geoalchemy-2.readthedocs.io

1

pamelafox.bsky.social @pamelafox.bsky.social · 2d

"If you love what we made... just love us, we made it!"

www.youtube.com/shorts/4DW3h...

They’re so mad that the Halftime show will be in Spanish

YouTube video by Josh Johnson

www.youtube.com

4

pamelafox.bsky.social @pamelafox.bsky.social · 2d

I'll be at the Pytorch conference in SF on Oct 22-23, hanging at the Microsoft booth in the afternoons. Stop by if you're attending!
events.linuxfoundation.org/pytorch-conf...

PyTorch Conference | LF Events

PyTorch Conference 2025: Join AI leaders, ML engineers & researchers in San Francisco, Oct 22-23. Experience the future of machine learning & deep learning.

events.linuxfoundation.org

1 3

pamelafox.bsky.social @pamelafox.bsky.social · 2d

🔴 Live @ 10am PT: "Retrieval Augmented Generation"
www.youtube.com/watch?v=DHeb...

Slides:
aka.ms/pythonai/sli...

Part 3 in our Python+AI series, join us!

Python + AI: Retrieval Augmented Generation

YouTube video by Microsoft Reactor

www.youtube.com

pamelafox.bsky.social @pamelafox.bsky.social · 2d

I only use Streamyard anymore, its surprisingly the most reliable software I've used.

pamelafox.bsky.social @pamelafox.bsky.social · 3d

LLMs are slow! But you can make the LLM answer appear faster by *streaming* the response, one token at a time.

Steps:
1) Call LLM API with stream=True
2) Return tokens over HTTP streaming response
3) Re-render in frontend

I wrote more in this post:
techcommunity.microsoft.com/blog/azurede...

The importance of streaming for LLM-powered chat applications | Microsoft Community Hub

Thanks to the popularity of chat-based interfaces like ChatGPT and GitHub Copilot, users have grown accustomed to getting answers conversationally. As a...

techcommunity.microsoft.com

3

pamelafox.bsky.social @pamelafox.bsky.social · 3d

🔴 Streaming at 10AM PT: Vector embeddings (Part 2 in our Python+AI series)
www.youtube.com/watch?v=ABLe...

Python + AI: Vector embeddings

YouTube video by Microsoft Reactor

www.youtube.com

2

pamelafox.bsky.social @pamelafox.bsky.social · 4d

Happy first-full-live-performance-of-Huntrix to those who celebrate:
www.youtube.com/watch?v=8y-V...

EJAE, Audrey Nuna and Rei Ami from KPop Demon Hunters: Golden | The Tonight Show

YouTube video by The Tonight Show Starring Jimmy Fallon

www.youtube.com

4

Reposted

Global AI Community @globalai.community · 4d

Tomorrow 6 PM at GitHub HQ! 🎙️

Valence AI: Emotion detection
Nox: Anticipatory AI
Pol Peiffer (Sierra AI): Voice agents

SF AI community gathers. Arrive early - limited space.

Last call ⬇️
aka.ms/aist-reg/gai...

#AIShowAndTell #Tomorrow

1

pamelafox.bsky.social @pamelafox.bsky.social · 4d

🔴 Live @ 9AM PT: "Large Language Models", the first stream in the Python+AI series

Watch: www.youtube.com/watch?v=X967...

Slides: aka.ms/pythonai/sli...

Python + AI: Large Language Models

YouTube video by Microsoft Reactor

www.youtube.com

1

pamelafox.bsky.social @pamelafox.bsky.social · 5d

Y si habla español, tendremos una serie para hispanohablantes, con @madebygps.com

🔗 Regístrese ahora:
aka.ms/PythonIA/Serie

pamelafox.bsky.social @pamelafox.bsky.social · 5d

My sister's an undergrad ML researcher and trying to choose a focus/lab. She's hearing conflicting advice about the best topics to pursue.
Any tips from those of you in the ML academia space?

pamelafox.bsky.social @pamelafox.bsky.social · 5d

Want to learn how to use Python with generative AI models?
Our free livestream series starts tomorrow!

For 3 weeks, every Tues/Wed/Thurs, tune in for:
🔍 LLMs, Vector Embeddings & RAG
🖼️ Multimodal Models
🤖 Agents & MCP
🛡️ AI Safety, Quality & Evaluation

Register @ aka.ms/PythonAI/ser...

Listing of Python+AI series dates and titles

1 1 8

pamelafox.bsky.social @pamelafox.bsky.social · 8d

awww and its a pseudo marquee!

1

pamelafox.bsky.social @pamelafox.bsky.social · 8d

yeah it took me a while too. you can PerksPlus it!

1 1

pamelafox.bsky.social @pamelafox.bsky.social · 8d

Yay see you there!

pamelafox.bsky.social @pamelafox.bsky.social · 8d

Currently reading "Empire of AI" by Karen Hao, a history of OpenAI based off interviews and documents.
bookshop.org/p/books/empi...

Really interesting to get more insight into how prioritization decisions were made. I now understand the OpenAI/Anthropic split better.

Empire of AI: Dreams and Nightmares in Sam Altman's OpenAI

Dreams and Nightmares in Sam Altman's OpenAI

bookshop.org

1 4

pamelafox.bsky.social @pamelafox.bsky.social · 9d

Yep I saw that! I'm asking on behalf of a dev who wrote their CLIs in Rust, so alas, he can't use that. Love that you have that for Textual though, as I write my CLIs in Textual usually.