this is f*cking insane, Claude Code agents turned a reading list into a lab you can run in your browser Boris Cherny (creator of Claude Code) gave 12 agents 64 research papers and one assignment: build a browser lab where readers can rerun the experiments pick a fluid simulation,... change the conditions and watch how the result changes switch methods while keeping the same scene. inspect the equations, code and measurements beside the animation give each group a specific job: → readers extract equations, assumptions and experimental settings → builders turn the methods into runnable code → reviewers compare outputs against the papers → interface agents build the controls for exploring each experiment a simulation can look convincing while producing the wrong answer reviewers check whether it preserves mass, stays stable and matches the paper's measurements before accepting the implementation every experiment has to show: > the claim being tested > the conditions used > the difference from the reference result > the details still needed to verify it the lab covered 11 methods across 28 reference scenes. seven matched the selected reference results within the stated tolerances; four stayed marked for further review give Claude Code your paper collection, then ask: "Turn these papers into an interactive browser lab 1. Extract implementable methods, equations and experimental settings. Attach page references and flag missing details 2. Assign independent methods to separate agents. Use shared inputs and measurement tools so the results can be compared 3. Have reviewers check each implementation against the paper. Record discrepancies and keep unverified results clearly marked 4. Build controls for changing parameters, comparing methods and opening the code behind each result" start with one paper and reproduce one reference experiment before adding more agents include the inputs, dependencies and checks so the next reader can rerun it from a clean checkout watch the agents at work belowshow more

i have no clue what this is, but it's so cool

basically a way to play with the experiments from a research paper

got it! thanks

Jarvis. Little bit more visuals and some more sloptext and our post is perfect. Make it sound and look cool so non technical people will go bananas

love the focus on reproducible results

makes it much more useful when someone else can rerun it and check the result

holyyy... sounds crazy man, will try today!

let me know which paper you try it on

Not only do vibe coders think they are talented, they create ridiculous visualisers to pretend something magical is happening. The fuck is wrong with these retards?

this is looks so good bruh

thank you <3

this is how i'd want to work through a paper

same here being able to change things and see what happens would help a lot

Visually appealing trash

@palehonk These visuals are lame, and what usefulness does this “lab” have, if any, other than giving people something to play around with?

lol more network graph animations impressing non-technical people. shit cracks me up

add Sogni Creative Agent for the explainer videos. keep the simulation in charge of the science 🫶

god tier performative key jingling holy shit

@mika_systems Wonder if they'll hit the data privacy wall soon. Managing that many agents with sensitive info is tricky.

@grok ok help me out what is going on in this post and video?

This is not good. 🤦🏿

12个agent跑64篇paper搭个能复现实验的浏览器lab…这个方向比整天吹'agent取代程序员'实在多了。说实话复现性才是科研的命根子，让读者能自己rerun比啥宣传都强🔥

isnt this slop tho?

Agents make it cheaper to turn papers into runnable experiments, but validation against reference results is still the deciding constraint, especially when some methods remain unverified.

Do you think it works for legal or judicial process tasks?

this example covers research simulations, i haven't seen it tested on legal tasks

i'd spend more time on the assumptions if i could mess with them instead of just nodding at the math

changing an assumption yourself would make it much harder to skim past it

Whenever I see a super elaborate page with exaggerated AI animation, I already know it's some kind of nonsense from a boastful charlatan.

safari will find a way to break it

Safari is the final boss lol

Выглядит как звезда 💫

this is the real fkng sanity

much easier to trust when you can actually rerun the experiment

@Thinker_View

@grok what do you think of this?

Look at that fancy interface!

12 agents and a reading list, wild

12 agents on 64 papers and none of them stepped on each other? want to see how he split the work

ah....rhea is the correct name....they figure that one out later.... 🤪

nb

Someone go have it figure out aliens