Quick re-up: First AI office hours session is today from 2-3pm ET! Already looking like a fantastic group with representation across over a dozen institutions. Register below for this one or future ones!

Brian Heseung Kim@brhkim.bsky.social · last wk.

I've run intensive workshops on best practices for responsible and rigorous AI adoption for well over 500 researchers+academics over the past few months. The Q&A after my workshops are some of the most fun and productive conversations I've had, so I figure: let's make this its own regular thing!

I've run intensive workshops on best practices for responsible and rigorous AI adoption for well over 500 researchers+academics over the past few months. The Q&A after my workshops are some of the most fun and productive conversations I've had, so I figure: let's make this its own regular thing!

🥳DAAF v3.0.0 is out today!! The two headliners: DAAF now speaks fluent R, and it can now ALSO run on your ChatGPT subscription/API key (Yes, GPT-5.6 inside Claude Code!). The whole release is about greatly expanding who DAAF is for across the social sciences. daafguide.substack.com/p/new-daaf-v...

Our @wheelockpolicybu.bsky.social group was thrilled to have @brhkim.bsky.social run workshops on AI for social science research. Many of our students/faculty have already gone from casual usage to deeper, more thoughtful engagement with the tools agentic AI now brings. Highly recommend!

AI can help us think *more* critically and *more* deeply, not less, and it's urgent that we start cultivating that practice for ourselves and our students. Lessons from a former English teacher battling SparkNotes: daafguide.substack.com/p/on-rethink... #EduSky #AIinEducation #HigherEd #AcademicSky

On rethinking coursework expectations in the era of AI: Raising the floor and the ceiling

What we can learn from high school English teachers and frontier AI model benchmarking

daafguide.substack.com

I've been running workshops on AI for academics across the country the past few months, and the same question has come up in every single room: So... what are we going to do about teaching? No easy answers, but I want to offer a few distilled suggestions: daafguide.substack.com/p/on-rethink...

On rethinking coursework expectations in the era of AI: Raising the floor and the ceiling

What we can learn from high school English teachers and frontier AI model benchmarking

daafguide.substack.com

Added GLM 5.2 to DAAFBench, and it's so good it required a complete rewrite of the key takeaways and recs. It is utterly indistinguishable from Opus 4.5/4.6/4.8 at 25-33% of the cost and open-weight. What a wild week. Kimi K2.7 Coder also added in, but frankly unimpressive!

Brian Heseung Kim@brhkim.bsky.social · 2mo ago

I ran 17 different frontier models through 51 different tests (2500 total runs!) to examine: How well do different AI models handle the complexities of rigorous quantitative data analysis workflows? Excited to introduce DAAFBench: Orchestration! daaf.openaugments.org/bench

I ran 17 different frontier models through 51 different tests (2500 total runs!) to examine: How well do different AI models handle the complexities of rigorous quantitative data analysis workflows? Excited to introduce DAAFBench: Orchestration! daaf.openaugments.org/bench

very cool use of claude code!

Brian Heseung Kim@brhkim.bsky.social · 2mo ago

🥳 If you’ve been wondering how to use Claude Code as a quantitative researcher/social scientist of any kind: I’ve *finally* made a very nice, very accessible, and very informative homepage for the Data Analyst Augmentation Framework (DAAF), and I think you'll wanna take a look!

🥳 If you’ve been wondering how to use Claude Code as a quantitative researcher/social scientist of any kind: I’ve *finally* made a very nice, very accessible, and very informative homepage for the Data Analyst Augmentation Framework (DAAF), and I think you'll wanna take a look!

🥳 DAAF v2.1.0 is live today! With it, I feel strongly that DAAF is now finally the best, safest, AND easiest (+free!) way to get started w/ Claude Code for *any* data work. v2.1.0 was focused on three core features for quality of life and ease of use: daafguide.substack.com/p/daaf-v210-...

DAAF v2.1.0: The Frictionless Update

In my very biased opinion: DAAF is now the best AND easiest way to get started using Claude Code for researchers

daafguide.substack.com

🙌 New explainer article! TLDR: Everyone needs to be investing in better logging+monitoring to track model adherence for their workflows, because we can't assume any two models will follow instructions the same way given how weird the AI frontier is right now daafguide.substack.com/p/opus-47-la...

The Opus 4.7 launch fiasco as a crucial reality check for anyone building with AI in 2026

Do you really know what your AI agents are doing right now?

daafguide.substack.com

A lot of peeps have asked: What does it actually look like to use DAAF to analyze data? And how is it better v. Claude Code alone? It's exactly the right Q, and so I put together this interactive walkthrough showing every step, doc, and output from a full project! openaugments.org/daaf_anatomy...

We’re excited to co-sponsor Machine Collaborators, a free conversation series led by Charles Crabtree on how researchers are using #AI in their work. Join the 1st session Thur, April 16 at 7pm ET (11pm UTC) as @brhkim.bsky.social discusses accelerating quantitative research w/ AI agents. Learn more:

Machine Collaborators

A global conversation series on what happens when researchers work with AI.

machinecollaborators.org

🥳 It's been a month since I launched DAAF, the Data Analyst Augmentation Framework... which means now is great time to celebrate the launch of DAAF v2.0.0 and re-introduce you all to a more useful, usable, and flexible tool for anyone analyzing data in their work!! www.youtube.com/watch?v=747r...

DAAF v2.0.0 -- Responsible, Rigorous, and Reproducible AI-empowered Data Analysis with Claude Code

YouTube video by Brian Heseung Kim (brhkim)

youtube.com

🥳 It's been a month since I launched DAAF, the Data Analyst Augmentation Framework... which means now is great time to celebrate the launch of DAAF v2.0.0 and re-introduce you all to a more useful, usable, and flexible tool for anyone analyzing data in their work!! www.youtube.com/watch?v=747r...

DAAF v2.0.0 -- Responsible, Rigorous, and Reproducible AI-empowered Data Analysis with Claude Code

YouTube video by Brian Heseung Kim (brhkim)

youtube.com

Very exciting: We've got over 150 registrants as of this morning!! If you're interested in getting caught up on the frontiers of using AI for rigorous quantitative research (not just education-specific), I'd love to see you there! #academicsky #academicchatter #econsky #phdsky

Brian Heseung Kim@brhkim.bsky.social · 4mo ago

LLM AI assistants will always be at risk of hallucinating/sycophancy/lying. Can they still be useful for accelerating good research? Yes! But we need a *lot* of guardrails. If you've wanted to learn how to use tools like Claude Code to *responsibly* accelerate quantitative research...

LLM AI assistants will always be at risk of hallucinating/sycophancy/lying. Can they still be useful for accelerating good research? Yes! But we need a *lot* of guardrails. If you've wanted to learn how to use tools like Claude Code to *responsibly* accelerate quantitative research...

As AI tools for research proliferate and research production accelerates, publishing is going to need to split into two tracks in the near future: that which is mechanically reproducible and verifiable, and basically everything else.

Excited to be spending the rest of the week at AEFP 2026! Lots of exciting opportunities to connect if you want to know where to find me: 1. Running a professional development session on Friday morn to help folks target job applications to non-research organizations for "embedded" researcher roles

Brazenly irresponsible projects like this are exactly what drove me to develop DAAF vocally centering the human expert. If we don't form strong, principled positions about what good AI-empowered research *can* and *should* look like, we wind up getting dragged around by people who clearly...

If AI tooling accelerates research production, how can our current publication systems possibly keep pace? One answer: AI will also allow us to rapidly verify the reproducibility of many claims, and we *must* modernize journals accordingly ASAP (Step 2) #AcademicSky #EduSky #ResearchSky

Brian Heseung Kim@brhkim.bsky.social · 5mo ago

How can we get to a more optimistic AI-empowered future for academia and science? Step 1: As AI allows us to accelerate the production of more and more scientific claims, we *must* center transparency and reproducibility as core requirements for all AI research tooling #academicchatter #econsky

How can we get to a more optimistic AI-empowered future for academia and science? Step 1: As AI allows us to accelerate the production of more and more scientific claims, we *must* center transparency and reproducibility as core requirements for all AI research tooling #academicchatter #econsky

Brian Heseung Kim@brhkim.bsky.social · 5mo ago

AI-empowered research is here, and it's going to break the peer review system. What can we actually do about it? daafguide.substack.com/p/a-better-a... #academicsky #highered #academicchatter #PhDChat #phdsky #econsky

AI-empowered research is here, and it's going to break the peer review system. What can we actually do about it? daafguide.substack.com/p/a-better-a... #academicsky #highered #academicchatter #PhDChat #phdsky #econsky

Academia is dead; Long live academia!

Six steps towards unlocking a more optimistic AI-empowered future for academia and science

daafguide.substack.com

New long-form explainer post! Why no one can agree about AI progress right now: A three-part mental model for making sense of this weird moment on the AI frontier daafguide.substack.com/p/ai-progres...

Why no one can agree about AI progress right now

A more useful and practical mental model for thinking about the AI capability frontier

daafguide.substack.com

Are you a data scientist/researcher trying to figure out how to get started with AI tools like Claude Code -- without sacrificing the transparency, rigor, or reproducibility you know *matters* for producing insights you can actually stand by?

DAAF Complete End-to-End AI Data Analysis Demo in 10 minutes

YouTube video by Brian Heseung Kim (brhkim)

youtu.be

Today, I’m launching DAAF, the Data Analyst Augmentation Framework: an open-source, extensible workflow for Claude Code that allows skilled researchers to rapidly scale their expertise and accelerate data analysis by 5-10x -- *without* sacrificing the transparency... github.com/brhkim/daaf/...

GitHub - DAAF-Contribution-Community/daaf: DAAF, the Data Analyst Augmentation Framework: An open-source, extensible workflow for Claude Code that allows skilled researchers to rapidly scale their exp...

DAAF, the Data Analyst Augmentation Framework: An open-source, extensible workflow for Claude Code that allows skilled researchers to rapidly scale their expertise and accelerate data analysis by a...

github.com

🚨Calling all education+social science researchers! 🚨 For the second time ever, Common App is hosting an open Call for Research Proposals to solicit innovative and rigorous research projects using our expansive (& still mostly untapped) data warehouse! www.commonapp.org/files/DAR/Co...

This tarriffs strategy feels like the foreign trade equivalent of injecting ourselves with bleach to fight an infection and so at least he's being consistent

Disappeared Tufts Human Dev. PhD student Rumesya Ozturk is also a graduate of Teachers College - she’s a dev. psychologist studying children’s media & prosocial development. She also bakes without recipes and binge-watches cartoons. She is our colleague & she was abducted on the street w/ our tax $.