AI Education | Developer Advocate @huggingface
EU
Joined November 2022
- Tweets1.2K
- Following562
- Followers4.4K
- Likes3.5K
Pinned Tweet
introducing tau τ — an educational agent harness that teaches you how to build agent harnesses
i will be publishing tutorials and demos on how to use it to create your own TUIs, harnesses, extensions, etc.
Happy Tau Day!! 🤓
👉 twotimespi.dev/
btw, here is the prompt that i used for this video.
"""
consider the documentation of pi-durable (present in this repository). create a motion graphics video that illustrates pi-durable's architecture and system design, and how to use it based on this documentation. feel free to use images from the internet if needed. i added the svg logo of pi to this folder for the video.
"""
the docs in question are my research notes about the package. no skills, no mcps. the model does a much better job with simple prompts.
i had opus 5.5 distill this 30 min lecture about pi-durable into a 7 min vid with my voice.
> qwen3 tts could do a better job cloning my voice
> the video looks great. i'm working on industrializing these kind of explanation videos.
good starting point, but i still recommend watching the full lecture for a more detailed explainer.
ds is the definition of openness. thank you
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
i had opus 5.5 distill this 30 min lecture about pi-durable into a 7 min vid with my voice.
> qwen3 tts could do a better job cloning my voice
> the video looks great. i'm working on industrializing these kind of explanation videos.
good starting point, but i still recommend watching the full lecture for a more detailed explainer.
Alejandro AO 🤗 retweeted
Entire industries will be built on open models.
Developers need the freedom to build them.
@JensenHuang on why keeping these models open and accessible matters.
i just recorded a full tutorial on pi durable
1:48 Checkpoints + crash recovery
8:38 Harness vs. app
13:46 App architecture + storage
19:30 Six setup steps
23:11 Local Qwen + llama.cpp demo + restart recovery
The demo runs Qwen3.5-0.8B with llama.cpp, reads a file using tools, then resumes interrupted work from SQLite.
Durability doesn't make arbitrary side effects safe to repeat. Tool replay safety matters.
github.com/alejandro-ao/pi-d…
Alejandro AO 🤗 retweeted
how do you fine-tune a decision model on your own data?
in these two notebooks, you'll learn:
• how to prepare the training data
• how to fine-tune our d1-3B decision model
• using @UnslothAI and @huggingface TRL
all within 30 minutes on a free-tier GPU.
unsloth: github.com/Liquid4All/cookbo…
TRL: github.com/Liquid4All/cookbo…
ps: how adorable are these halloween kitties? 👻
the spark is open 🤗
Muse Spark Open -- weights releasing today -- runs great on new Mac Studios thanks to work with @johnternus and Apple.
really fun! i did collabs with @MilksandMatcha when she was at exa, and it was amazing. great podcast!
“Do you think that getting married young has made you less ambitious?”
Welcome to the first episode of SOTA Wives with Sarah Chieng (Head of DevX at Cerebras, @MilksandMatcha) x Vivian Midha Shen (General Partner at Y Combinator, @vivianmshen).
In episode 1: finding your person before getting rich, turning down startup funding to finish MIT, what a YC partner sees in founders, and how a birthday post became “psychological violence.”
4:08 Finding your husband before the exit
8:48 Did we marry away our ambition?
18:04 The internet called Sarah's birthday “psychological violence”
20:47 Does your YC partner secretly think you’re doomed?
23:47 Building a startup with your ex-boyfriend
28:54 SF cat parties
pro tip: market to the 2nd group -- 2B people who have used AI, but haven't had their mind blown in a while
Alejandro AO 🤗 retweeted
I turned @googlegemma EmbeddingGemma 2, an embedding model, into a speech-to-text model, and it turned out to be one of the fastest STT models you can run on device.
On a base M3 MacBook Air it transcribes an hour of audio in about 13 seconds: 288× real time.
The idea came in March with Gemini Embedding 2. It puts text, images, audio and video into one shared space, so a spoken word lands right next to the written word. If that's true, you should be able to read a transcript straight out of the embedding model. I couldnt tried it then, but last week EmbeddingGemma 2 came out as an open model, so I tried it.
this is true. i feel both relief and melancholy for this shift in how we build.
