Tired: Skill.MD

Wired: Dr. Skill

Aspired: Skill.MDPhD

Bird’s eye view of Stanford

Statisticians: “Data” is plural but “dataset” (a set of multiple data) is singular. Thank you for your attention to this matter.

House full on amazing panel discussion on Clinical Reasoning in the Era of AI at CDI2 conference at UCI @kdpsingh.bsky.social

GitHub - rohit43/oneBit Contribute to rohit43/oneBit development by creating an account on GitHub.

1 bit LLM is crazy!
How on the earth they pulled this off??

github.com/rohit43/oneBit

If you work with UC Health data or aspire to do so then it might be worth attending the CDI2 conference at UCI

na.eventscloud.com/website/5224...

Coffee with Blue jay at Big Sur

Claude Mythos system card is quite something!

www-cdn.anthropic.com/08ab91580709...

Bacha hai tu yaar mera, chal tujhe Fanta pilata hu! 😂

Tired: Reasoning
Wired: Latent reasoning

Great work! Congratulations 🎉

GitHub - rohit43/oneBit Contribute to rohit43/oneBit development by creating an account on GitHub.

PrismML 1-bit 8B LLM!! 👀
prismml.com

I tried and it’s not only fast but also generate somewhat okay looking answer to the question - ‘What is Causal Inference?’

1-bit compression/training is insane

github.com/rohit43/oneBit

FDA Draft Guidance on Bayesian Methods in Trials This Perspective discusses the use of bayesian methods in clinical trials and the importance of having US Food and Drug Administration (FDA) guidance that provides substantive insights about the metho...

3. Reflections on FDA Draft Guidance on Bayesian Methods in Trials—Protecting Scientific Integrity and Evidentiary Standards: jamanetwork.com/journals/jam...

FDA Draft Guidance for the Use of Bayesian Methods in Clinical Trials This Perspective discusses the draft guidance for industry that the US Food and Drug Administration released on the use of bayesian methods in testing therapeutics.

2. FDA Draft Guidance for the Use of Bayesian Methods in Clinical Trials: jamanetwork.com/journals/jam...

Embracing Bayesian Methods in Clinical Trials This Perspective discusses the importance of the US Food and Drug Administration’s draft guidance on the use of bayesian methods in clinical trials because it underscores its commitment to modernizing...

In case you looking for a break in the news cycle.

In statistics news, back to back in JAMA

1. Embracing Bayesian Methods in Clinical Trials url: jamanetwork.com/journals/jam...

TurboQuant: Redefining AI efficiency with extreme compression

This is good, like really good.

There’s so much efficiency hidden in those vector spaces!

research.google/blog/turboqu...

Got fully distracted today :-)

Tired: Layers
Wired: Semantic layers

Gemini Embedding 2: Our first natively multimodal embedding model An overview of Gemini Embedding 2, our first fully multimodal embedding model that maps text, images, video, audio and documents into a single space.

LOTR of embeddings? Where R is a modality ring!

blog.google/innovation-a...

Tired: Storage
Wired: Agent friendly storage

Appreciate the years of work by Andrew Wong, Yasir Tarabichi, Pat Lyons, and colleagues around the country to understand how routinely implemented predictive AI tools perform in diverse real-world clinical settings.

Here’s a quick timeline...

Donald Knuth on Claude.

Wow!

Time for Knuth to send ‘Knuth reward check’ to Claude :-)

www-cs-faculty.stanford.edu/~knuth/paper...

Qwen3.5 - a Qwen Collection We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Small language models are the new large language models. This is amazing! Will taking them for a spin huggingface.co/collections/...

It’s all the more important to build local, on device language models from scratch as well as enable synthesizing health system scale data. Excited to discuss some of these efforts (UCSF-GPT, & diveEHR) at upcoming UCSF AI day!

Happy to chat, & connect if you are around.

ai.ucsf.edu/researchday

/simplify

New plugin from #claudeCode is epic, especially if you are learning best practices of how to structure your thoughts into a good looking, optimal, and readable code.

GitHub - rohit43/agenticOMOP Contribute to rohit43/agenticOMOP development by creating an account on GitHub.

Pre-paid API plan with Claude code is also very optimal. It feels like pre-paid phone recharge :-)

Like it just barely any tokens to achieve this task github.com/rohit43/agen...

Recent access to ChatGPT enterprise edition only made me realize how good Claude training has been overall.

Agentic #OMOP experiment: dusted off some of my old scripts and let ClaudeAI orchestrate everything into a full OMOP data report. This is kind of crazy. 🤯

Code: github.com/rohit43/agen...

JCHI Executive Director @kdpsingh.bsky.social
& co-host Navdeep Tangri explore AI’s evolving role in nephrology in “Rewired: Navigating AI's Role in Kidney Health," a new podcast from @asnkidney.bsky.social.

Listen ⬇️
t.co/YMYjX7PbmQ