iiinova — where your digital self keeps evolving

Brian Heseung Kim @brhkim.bsky.social · 08/09/2026

Last time, we had a phenomenal convo about the role of fully "local" offline models for use with private/sensitive data, as well as the value of using a *variety* of AI models in tandem to balance capabilities and costs against the difficulty of the task Register here: forms.gle/E7UF5r2KFMoL...

forms.gle

AI Office Hours with Brian Heseung Kim/Open Augments: Registration & Questions

Free, biweekly, one-hour open discussion/troubleshooting/ask-me-anything sessions for all things AI on Google Meet with Brian Heseung Kim (Founder and Chief Data Scientist, Open Augments: https://open...

000

Brian Heseung Kim @brhkim.bsky.social · 08/09/2026

So what's the deal with GPT-6 Astra/Fable 5.1? What is "computer use" and why is it such a big deal right now? Are we at AGI/RSI?? Swing by for another round of AI Office Hours this Thursday, 2-3pm ET to learn more about all of the above (and anything else you bring) as the field continues to shift

100

Brian Heseung Kim @brhkim.bsky.social · 03/09/2026

If you’re not a part of APPAM, they’re my second professional home and one of the most fantastic community of policy researchers out there. Can’t recommend enough getting involved, especially given their proactivity and leadership in this space!

000

Brian Heseung Kim @brhkim.bsky.social · 03/09/2026

(2) A pre-conference office hours-style webinar specifically for researchers to come and ask anything about AI, and... (3) Open drop-in consulting and office hours throughout the entirety of the conference for more bespoke troubleshooting and coaching (free for attendees!)

100

Brian Heseung Kim @brhkim.bsky.social · 03/09/2026

Very, very excited about this collaboration: If you’re an Association for Public Policy Analysis & Management member and/or planning to attend their Fall Conference, I’m going to be offering: (1) My intro series of workshops for attendees on responsible AI for policy research...

121

Brian Heseung Kim @brhkim.bsky.social · 02/09/2026

Watch Lesson 3 from the Open Augments AI Academy course map (or start with Lesson 1 if you're new!) linked above, and then practice everything hands-on in the Context Gym. New exercises for this lesson, free as always in your browser: openaugments.org/academy/gym

openaugments.org

Open Augments Context Gym - A Hands-on Practice Floor for AI

The AI Academy's interactive practice space. See how system prompts, memory files, tools, and more shape how AI assistants think and work. Free, forever, for everyone.

000

Brian Heseung Kim @brhkim.bsky.social · 02/09/2026

This course is meant to catch people up at light-speed on AI, 2023 to now. Lesson 1 got you to ~2024. Lesson 2 (prompt engineering) to early 2025. Today's lesson -- tools and agents -- gets you caught up to the top of 2026 in concept. And we're still just getting started!

Video thumbnail for Lesson 3 of the Open Augments AI Academy: Brian smiling and holding up a red hammer next to the text "AI + TOOLS = LIMITLESS," with phone, document, and browser app windows clustered below, on a dark navy background with a teal glow.

100

Brian Heseung Kim @brhkim.bsky.social · 02/09/2026

Why does it feel like AI assistants can suddenly do... everything right now? Lesson 3 of my fully-free AI Academy is out: the biggest buzzwords of 2026 (agents, tools, harnesses, context engineering) demystified in plain English, in 18 minutes flat! openaugments.org/academy#l1-05

openaugments.org

Open Augments AI Academy - A Free, Interactive AI Course for Everyone

A fully-free, interactive course on using AI responsibly and effectively. Short weekly video lessons and hands-on demos -- no jargon and no experience required.

111

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

I genuinely don’t think there’s a better way to get caught up (that’s why I’m making this: for free, forever, for everyone!), and I hope you’ll take the time to dive in!

000

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

From there, we’ll get everything practiced and really digested through the end of Level 1, and Level 2 is where we start getting through the advances of 2026. I’m trying to make this the smoothest yet *accelerated* on-ramp I can for anyone who wants to genuinely understand this moment

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

Lesson 1 catches you up to where the AI world stood in ~2024. Today’s lesson on prompt engineering gets you to the beginning of 2025. Next Wednesday’s lesson -- on tools and agents and context engineering -- catches you up to roughly the end of 2025.

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

And there are 8 new hands-on Gym exercises accompanying today's lesson (auto-compaction, memory files, custom system prompts, etc). We are still just getting started. One way to think about it: My goal is that this series is a crash course to catch people up on these crazy past two years in AI.

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

If you're just hearing about the AI Academy: I strongly recommend starting with Lesson 1 and running through its practice exercises in the interactive Context Gym first -- this lesson builds directly on both. openaugments.org/academy/gym

openaugments.org

Open Augments Context Gym - A Hands-on Practice Floor for AI

The AI Academy's interactive practice space. See how system prompts, memory files, tools, and more shape how AI assistants think and work. Free, forever, for everyone.

200

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

- Why more context is NOT always better: two extremely common failure modes called context rot and auto-compaction that explain an enormous amount of everyday "why did my AI suddenly get dumb?" frustration -- and one simple habit you can start practicing that prevents them outright

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

- How to start "loading the dice" on these autocomplete engines to get dramatically more useful and reliable responses back (what people call prompt engineering)

200

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

In this 18 minute lesson, you'll learn: - The most practically useful mental model in the entire course: thinking of your AI assistant as having two very different kinds of memory (one fuzzy and unreliable "long-term" memory, one sharp and steerable "short-term" memory)

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

Lesson 1 covered how these AI assistants work under the hood (autocomplete with an extremely fancy hat on). While this is a fatal flaw, this lesson is where we learn how to start making that flaw work *for* you as its greatest strength, as well: endless flexibility and adaptability.

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

These videos are a LOT to put together at the level of production quality and accessibility I'm shooting for, and I'm so grateful it's resonating with folks (things to remember when I'm frantically editing videos with Claude past midnight!!). openaugments.org/academy#l1-04

openaugments.org

Open Augments AI Academy - A Free, Interactive AI Course for Everyone

A fully-free, interactive course on using AI responsibly and effectively. Short weekly video lessons and hands-on demos -- no jargon and no experience required.

100

Brian Heseung Kim @brhkim.bsky.social · 26/08/2026

Lesson 2 of the fully-free Open Augments AI Academy is out today: "How to Make AI Less Bad: Context Is Everything" First, HUGE thanks to folks for sharing and digging in: the course launched just a week ago, and we're on track break a thousand subscribers on YouTube today or tomorrow. W I L D.

100

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

The video above is a short 2min pitch, but the full 10min course overview and the first lesson (16min) are live right now on the Course Overview page. Start there, and if you find it helpful, please do Subscribe and share with friends as I release lessons weekly! openaugments.org/academy

openaugments.org

Open Augments AI Academy - A Free, Interactive AI Course for Everyone

A fully-free, interactive course on using AI responsibly and effectively. Short weekly video lessons and hands-on demos -- no jargon and no experience required.

100

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

so that you, your coworkers, your friends, and your mom can follow along. I pair these videos with hands-on demos and interactive activities in the "Context Gym:" my way of giving you a safe and guided place to practice some of the core principles that should deepen your intuition as we go.

100

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

From there, we're going to learn about all the crazy buzzwords (context engineering, harness engineering, and Agents, oh my!) and advanced techniques, with much more to come! No jargon, no experience required, and all taught with the care of a former high school English teacher

110

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

When you really understand what's happening under the hood and how it works (no math or stats required!) a lot of confusing AI behavior suddenly clicks: why it hallucinates, why it's sometimes confidently wrong, and why it can now do way, WAY more than just write words on a page.

110

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

The first lesson starts with one foundational idea that most people miss when they get started with AI: that modern AI is much less like a hyper-intelligent database or brain, and much more like autocomplete with an extremely fancy hat on. That's its greatest flaw *AND* its greatest strength in one

110

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

As someone who's been using these tools and their predecessors for my research since ~2019, I'm trying to provide *everyone* the intuition and critical awareness they need to get started at this very confusing and pivotal time.

110

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

I'm calling it the Open Augments AI Academy. It's built for anyone out there who's seen all the crazy hype and discourse around modern AI and is just looking for a guided, grounded, and sane way to move forward learning how to approach these tools for themselves.

110

Brian Heseung Kim @brhkim.bsky.social · 19/08/2026

Over the past six months, I've been teaching teams at places like Stanford, Penn, Northwestern, and many more how to start using AI responsibly+effectively in their work. Today, I'm starting to release my entire curriculum: for free, forever, for everyone!

1115

Brian Heseung Kim @brhkim.bsky.social · 10/08/2026

Re-upping for the group: Next session is this Thursday, August 13th! We've got folks across institutions and industries, and we had a great last session talking through things like getting Claude to run more careful and rigorous web searching on your behalf Would love to see you there. Swing on by!

000

Brian Heseung Kim @brhkim.bsky.social · 30/07/2026

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!

000

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

Whether you've been to one of my workshops or webinars and left with a question you never got to ask, or you're deep in the weeds on agentic orchestration and want to learn some pro-tips, or you've just been seeing my posts and wondering what the heck all this AI stuff is about: this is for you!

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

- You can submit questions/topics ahead of time when you register, so I can come prepared with real answers (and real demos, where useful). But totally optional -- registration takes two minutes just to grab contact info (which will always be kept private and protected).

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

- Every level of experience is welcome. The "embarrassing" beginner questions are always worth discussing in more detail, and I can guarantee you're not alone.

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

- This is for kind of everyone! Most of my experience is supporting researchers, academics, and non-profits, but if you're interested and in another field or role, no harm in stopping by and joining the conversation.

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

- It's genuinely free, and always will be, with no catches or pitches. I am dedicating 40% of my time to pro bono work via Open Augments, and I love teaching. I want it accessible to anyone who wants to think more critically and carefully about AI given all the crazy hype/misinfo floating about.

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

Drop-in, drop-out, attend one, attend all, talk a lot, or just listen in off-camera. Taking all kinds, because there's always more to learn from others on this crazy and wild frontier right now! A few things I want to emphasize:

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

Bring your questions, half-working pipelines, skepticism, worries, and ethical quandaries. No question too basic, too weird, or too heady/philosophical. I can't promise satisfying answers to *everything*, but I can promise helpful thinking and opportunities to dig in deeper to learn more together

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

That in mind, I'm excited to launch a nice little community service, free and open to the public, for as long as it continues to feel useful to folks: Every other Thursday from 2-3pm ET, I'm hosting free AI Office Hours: an open session for folks navigating AI in research/education/teaching/etc.

100

Brian Heseung Kim @brhkim.bsky.social · 28/07/2026

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!

110

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

Whether you’re one of the ~2,000 unique installers of DAAF so far or just getting started today, I’d love to hear from you!! Learn more about DAAF more generally and what it has to offer at our home page here: daaf.openaugments.org

daaf.openaugments.org

DAAF - Data Analyst Augmentation Framework

A free, open-source AI toolkit for rigorous research. DAAF helps skilled researchers deepen and extend their expertise with Claude Code -- while enhancing transparency, rigor, and reproducibility.

010

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

We must take an active role in shaping what rigorous, responsible research looks like in this new paradigm. We cannot afford to cede the territory to people more than happy to fill the void with “research-shaped objects” and slop. I hope DAAF can serve as one resource of many in that direction!

100

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

I remain more firmly convinced than ever that these tools have an immense transformative potential for the scientific community, and our society writ large by proxy… but if and only if we as trained researchers and experts are critically aware and informed consumers and contributors to these efforts

110

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

And the thing I've been promising forever: a full 30-minute install tutorial for researchers who've never touched Claude Code, Docker, or a terminal. Video: youtu.be/BPlR9bXZxnY And the Get Started install guide: daaf.openaugments.org/get-started.... #RStats #DataScience #OpenSource

youtu.be

Claude Code for Social Scientists: Get Started with DAAF in 30 Minutes

YouTube video by Brian Heseung Kim (brhkim)

120

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

Also new: run DAAF on Anthropic, OpenAI, or OpenRouter (Kimi K3, GLM 5.2, and friends), one friendly control panel instead of scattered helper scripts, a self-healing updater, and as many side-by-side installs as you like. Free and open-source, as always.

100

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

For the R folks (social science, ed policy, epi): R is now a first-class execution language; switching from the Python default takes one sentence. And for PII/HIPAA/FERPA-bound data: a new workflow helps you prototype your analysis on a fully synthetic "twin" dataset. The real data is never touched!

100

Brian Heseung Kim @brhkim.bsky.social · 17/07/2026

🥳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...

151

Reposted by Brian Heseung Kim

Joshua Goodman @joshua-goodman.com · 13/07/2026

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!

083