ieuanking

In this complex task, a multi-hop prompt instructs the agent to go through a 8 newly imported PDFs (stored locally on my desktop) to find important quotes, claims, and points made by the authors that could be used in a paper im working on.

After finding these points throughout the 8 files, the agent should create a new document that cites findings in-text, with verifiable click through.

With Gemini 3.1 Flash, Ubik Agents analyzed 8 PDFs, found 49 usable quotes, and generated a 1000 word annotated guide that explains its findings with in-text citations that verifiably link to cited points of information across documents inside my workspace in 6 minutes (video 2x speed).

Try Ubik: www.ubik.studio/download

yeah i took down :/ The password is a natural force, often caused by water, that reshapes landscapes.

A non-linear essay on technology, my childhood, and where we're headed.

I build AI tools for knowledge work and am a multi-medium digital artist.

More work/links here: https://aka4uh.com/ <3

God doesn't roll dice, God roleplays.

Disappointed, astonished, and somehow unsurprised.

Ubik is free until 2026! While we figure out how to make Ubik the best research AI tool possible, we value our users' feedback instead of churning quickly in these early days. This design philosophy has really helped us develop with the user in mind. That being said we plan to have a Free, Pro ($12-$15 + pay as you go with model usage), and Enterprise. If you have any feedback, please let me know! Happy you enjoyed Ubik. You can reach me at ieuan@ubik.studio any time :)

I've built a similar platform with searching access to arXiv and Semantic Scholar; the only difference is that our agents can highlight text in the paper down to the line level. In our testing, Gemini struggles compared to Sonnet four or Opus 4. We found that without agentic highlighting, there wasn't much difference in output quality or utility (meaning, references saved, generating with citation, or even quote gathering is still hard without actual PDF interactivity). I'd love your feedback on https://www.ubik.studio (use academic search)

I was just making a light acid joke :/ lmao didn’t mean to ruffle feathers

<300 blotter will make anyone artificially intelligent. Brain loops are scary; maybe AI models are just trapped in psychosis.

tysm!! means a lot <3

How will the HumanAlarm knockers wake up for their bookings? Will they book other HumanAlarms?

Also, you do not need to sign up to use Ubik - I am fixing that copy right now :/

Forgot to mention:

Our academic search uses ArXiv and Semantic Scholar, so when using the academic agent, you can convert open-access papers into AI-ready docs directly from the databases. <33

The video is sped up at 2.5x, and the actual time spent was around 7 minutes.

We are now setting up an Eval suite and hope to run evals on our agents within the coming month. We are almost finished with the entire local app.

Ubik is an AI studio with two agents: Fiig (built for research workflows (searching, analyzing, and writing)and Precog (Just like Fiig, but Web search-based instead of using academic databases).

If this looks like a helpful tool for you, I'd love to hear your feedback!

try ubik -- https://app.ubik.studio/chat

Integrating technology into classroom instruction has become inevitable, yet many educators lack sufficient training and access to appropriate tools. This disparity between administrative expectations, teacher preparedness, and student technological proficiency creates a significant imbalance in learning environments. While students are often well-versed in using digital tools, teachers frequently report difficulty effectively utilizing these resources for pedagogical purposes.

Sorry about that, will do - can't edit now :/ Shared on a busy train coming home from work <3

I'll definitely give this a listen, tysm!

not me realizing all my friends are autistic

Yeah, watched this one! I love her stuff. Her book Lost In Math really stuck with me over the last couple of months

lmao super fair - I just always see so much ahte on her, but I love her books and think she is a great science communicator.

I surround myself with people I trust. They tell me when I'm wrong and when I'm right. Everyone is right, everyone is wrong, and everything has bias. Instead of trying to discern what is real and fake, I think you should try to decide what you believe, and through that belief, you will find liars and trustworthy people full of valuable information. Most importantly, try to understand people who disagree with you - I grew up with a family with entirely different values and beliefs than I do. Having people around you who agree and disagree with you, with whom you feel comfortable discussing those topics, is vital to personal development and growth.

Thank you so much! PKD is the GOAT. Glad my dad introduced me to his writing when I was little, it has helped me laugh, escape, and analyze life through a lens I'm truly grateful for.

Even though my favorite games this last year have been built by indie devs, I still think fewer shortcuts shouldn't be met with reward. Like, "vibe [xyz]" and "innovating imagination" are just dog whistles that scream: Idk what I'm doing.

I find it disturbing how little safeguarding there is around ChatGPT and preventing these relationships (I develop AI Agents and have a small startup). I honestly worry so much for today's kids. How they will act without readily available answers and parasocial relationships -- it reminds me of how iPad babies form an addiction to the screen (screen dependency) that can be as severe as drug addiction (as a 2000 baby, I feel lucky to be more addicted to cigarettes than my screen). Somehow, I feel healthier and more human.

We didn't have smartphones as 4-year-olds. I got my first smartphone at 13, and the internet wasn't just a commodified hellhole yet. I loved to go outside and hang out with friends. It's crazy that the top level of screen addiction was video games (which come on, video games are fire, and if anything can be more educational and skill-building than ChatGPT).

With all this new AI school technology, I can’t help but compare the language around AI education to the “benefits” of the Smart Board, which helped nobody and, in the end, ironically made teachers look dumb.

The similarities between the universe beginnings and the formation of black holes, is to me, veryyyyy similar.

As a musician I love this especially because i constantly working with files (WAVs) would be cool if you could do some actual file browsing inside the drop down + like with music I want to be able to listen ot the wav before adding quick to rekordbox or ableton. Even for work (im a front-end dev for AI research platform) I think it would be helpful to do more browsing rather then selecting and then working from within that selected folder.

I am not a physicist- I am an anthropologist and front-end developer of an AI research studio I started with my best friend.

I love to learn and became interested in physics and astrophysics, but with a history in religious anthropology, I started making some connections I wanted to share.

If I got anything wrong, please tell me. These are some excerpts from a larger project I'm working on about Science and Religion and how faith in answers ends in a joint pursuit for parsimony.

Just in the sense that like -- everyone starts somewhere, don't glorify that process with the word vibe, reminds me so much of "Smart Boards," the dumbest technology to ever hit the classroom. Same with vibe physics or any of these new "vibe" hobbies, its great to be interested, worrying if taken 2 seriously because vibe=guess

Yeah, for sure, something I've come to realize as a 25 y/o AI dev & researcher is that effective AI use requires effective learning first -- vibe coding is great but only from a learning perspective, feels like we've just put some kind of value on the process of learning -- idk kinda strange 2 me