Hacker News Reader

samstaveon Strobelight: A profiling service built on open source technology

DOPE

Fractal compute expense modeelling is hard.

One may do well in applying fluid dynamics (such that we cannot maintain in head)

into compute requirements, it will be funny once we realize that everything i mico (pico) fluid dynamics in general

samstaveon Ladder: Self-improving LLMs through recursive problem decomposition

It reads like the famous Churchill quote about "if you gave me poison I would drink it"

samstaveon Ladder: Self-improving LLMs through recursive problem decomposition

F_yes

The thing to take here is that this should be a function_callable *feature* of a bot.

Basically, when architecting a persona ; "USE THESE RULES OF ENGAGEMENT"

Also -- where my CheckListManifesto folks at -- While we are building Patterns/Personas/Purgatories for our bots... We need to be able to reference a central CODEX of :

"Do this task but imbue yourself with (XYZ) name places"

AND LEARN FROM THE OTHERS

(so maybe a task marketplace of AI persona action?)

@ callable in an IDE

samstaveon Ladder: Self-improving LLMs through recursive problem decomposition

THe person's view to be scared of is peter thiel... hes the digital version of gorge soros. Meaning that this guys legacy is vast... and we have no clue how it will manifest over decades (especially after he is dead) -- but peter thiel is the leviathan of the digital future.

he is scary AF. he basically weaponized what george soros was but is still active.

samstaveon Things we've learned about building products

(Sorry, forgot to mention: trillions in tons of plastic E_waste.exe __ obsolete phone cases ==

F IVE

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

I watched them...

So amazed how aligned we are.

samstaveon Tailscale is pretty useful

Imagine a network of tailscale droplets that allow for agents to exit via a tailscale node in their locale and do work - then funnel data back to master wherever...

(Like If I wanted to crawl an area in germany, but needed an exit tracert that originated there - a tailscale droplet that could be connected to, perform [object] and openly pipe data back to master?

samstaveon Things we've learned about building products

The other issue with A/B is the imbued bias that comes from [TEAM] on [THING]

Meaning TEAM A has a want, and TEAM B has a want.. and its not about best feature its about "I told you so"

AB testing is great for sussing out use-flow-patterns, but oft may be used for ego (design team) testing...

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

Yes, as I am the one who wrote that.

samstaveon Apple M3 Ultra

Yes, but now its in my pocket.

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

I'm an idiot...

anyway - I just watched it.

Fantastic.

---

What are your thoughts on API-verload -- We will need bots and bots to keep track of the APIS...

Meaning... Are we to establish an Internet Protocol ~~ AI Protocol addressing schema - whereby every API henceforth has an IP v6 addy?

and instead of "domain names" for sale -- you are purchasing and registering an IPv6-->API.MY.SERVICE

samstaveon AI Will Upend a Basic Assumption About How Companies Are Organized

If you boil to the essence;

AI will reveal true desire, passion, resolve in any cluster of conscience.

Meaning... either an individual or group or organization will thrive... (and thats it)

(but the good, bad and ugly... at min it is the litmus of our current time)

samstaveon Show HN: Bayleaf – Building a low-profile wireless split keyboard

Beautiful!

I have these ultra thin bt keyboards that came in a leather ipad case from Restoration Hardware... SUPER THIN.

These bitches better have a lot of magnets! (oh saw that in future. and the LED thingy -- if you can have a display in the side panels, that would be nifty - and a second BT object? Whereby the display in the side panels is BT to phone and thus get distractions.

my problem with small super thin BT keyboards is that the materials are too slippy -- so a tacky-ish finish so they dont slide around when on a surface such as a pant leg or so...

also, a curved version of this would be great for a sleave version - with magnets so that you slap one on your fore-arm. could even have power/connectivity coming from wires in the garment that they attach to.

Magnets basically. Need lots.

what if... since the thing is so small, the actual object is the mouse? So you can choose l/r object and then just grab it and mouse about?

samstaveon Tailscale is pretty useful

> I can also access my home network easily.

expand plz

samstaveon QwQ-32B: Embracing the Power of Reinforcement Learning

>>In the initial stage, we scale RL specifically for math and coding tasks. Rather than relying on traditional reward models, we utilized an accuracy verifier for math problems to ensure the correctness of final solutions and a code execution server to assess whether the generated codes successfully pass predefined test cases

They should call this the siphon/sifter model of RL.

You siphon only the initial domains, then sift to the solution....

samstaveon QwQ-32B: Embracing the Power of Reinforcement Learning

>I love that emphasizing math learning and coding leads to general reasoning skills

Its only logical.

samstaveon Apple M3 Ultra

WHen I was with Mirantis, I flew to Austin TX to meet a client in a non-descript multi-tenant office building...

we walked in and getting our bearings, we come upon CRAY office. WTF?!

I tried the doors, locked - and it was clearly empty... but damn did I want to steal their office door signage.

samstaveon Apple M3 Ultra

"unified memory"

funny that people think this is so new, when CRAY had Global Heap eons ago...

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

Basically agentic personas or modus operandai

You tell the agent “grok this persona to accomplish the task

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

Dub, yeah I saw that but hadn’t listened yet.

What I want is a “podcast” with audience participation..

The lex fridman DeepSeek episode was so awesome but I have so many questions and I get exceedingly frustrated when lex doesn’t ask what may seem obv to us HNers…

Back topic:

Reports are flat; dossiers re malleable.

As I mentioned my goal is fractal visuals (in minamGL) of the true entanglements from the WEF out.

Much like mike Benz on usAid - using jina deep research, extraction etc will pull back the veil on the truth of the globalist agenda seeking control and will reveal true relationships, loyalties and connections.

It been running through my head for decades and I finally feel that jina is a tool that can start to reveal what myself and so many others can plainly see but can’t verify.

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

So I’ve started a thing with Jim’s, and the first effort I am doing is erring the “tone” meaning I’m building a project template that will keep the bots focused.

I think that one part of the deep loop needs to be a check-in on expectations and goals…

So instead of throwing a deep task: I find that bots work better in small iterative chucks of objectives..

I haven’t formulated it completely yet but as an example ive been working extensively with cursors whole anthropic abstraction ai as a service:

So many folks suffer from “generating” quagmire;

And I found that telling the bot to “break any response into smaller chunks to avoid context limitations” works incredibly well…

So when my scaffold is complete the goal is to use Fabric Patterns for nursery assignments to the deep bots.. whereby they constantly check in.

Prior to “deep” things I found this to work really well by telling the bots about obsessively development_diray.md and .json tracking of actions (even still their memory is super small, and I envisioned a multi layer of agents where the initial agents actions feed the context of agents who follow along and you have a waterfall of context between agents so as to avoid context loss on super deep iterative research…

(I’ll type out something more salient when I have a KVM…

(But I hope that doesn’t sound stupid)

samstaveon The Differences Between Deep Research, Deep Research, and Deep Research

First, SimonW, I devour everything you write and appreciate you most in the AI community and recommend you from 0 all the way to 1!!!

Thank you.

Second, thank you for bringin up Jina, I recently discovered it and immediate began building a Thing based on it:

I want to use its functions to ferret-out all the entanglements from the roster from the WEF Leadership roster, similar to the NGO fractal connections - I’m doing that with every WEF member, through to Congress.

I would truly wish to work with you on such, I so inclined..

I prefer to build “dossiers” rather than reports, and represented in json schemas

I’m on mobile so will provide more details when at machine…

Looping through a dossier of connections is much more thoughtful than a “report” imo.

I need to see you on someone’s podcast, else you and I should make one!

samstaveon Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

You FRiendly Spy drops a grain of rice in a plant... Pull up that AINT

samstaveon Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

This exposes what truely sysads did.. but in millions of clicks and interactions rather than bullet points.

THIS is what AI was SUPPOSED to take out of employment...

samstaveon Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

If you guys arent goofing around with such requests of your bots??

Are mine Sophomoric??

(aside from deep science/law/poli whatever..

arent you asking you bots to X10++??

(X10 as in the first-ish attempt a home automation)

I truly wonder where those folks went)

samstaveon Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

https://i.imgur.com/c5zjhlr.png

---

https://i.imgur.com/VxGwwPo.png

____

WE NEED GAURDRAILS NOW.

(An this is from me trying to crawl the darn WEF...

---

Thoughts appreciated

https://i.imgur.com/AJB3Mr0.png

https://i.imgur.com/59eD7o4.png

(what a fun rabbit hole this is

samstaveon Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

https://i.imgur.com/nynlP0o.png

samstaveon Firefly ‘Blue Ghost’ lunar lander touches down on the moon

I was wondering that as well as the crater edges look smoothed... Billions of years of Space Dust? (Obviously a fresh impact crater would be starkly different, but if its craters are all from after it was formed, and then was still being inundated with a cloud of micro hits - and then the cloud of all the dust that was left from all the collisions turned to that powder, and finally settled over the surface.

samstaveon Firefly ‘Blue Ghost’ lunar lander touches down on the moon

Yeah, CUDAs to the Team!

samstaveon A Letter to the American People

It would be lovely if you could have the USPS scan and send all physical mail electronically (as a paid Mail + Pro fee) --and have it piped directly to a GPT who reads all the mail, logs the data into Sqlite3 and then have summaries, reminders, due dates, amount, budget hits, as daily summaries from your GPT of choice..

And then you can block marketing mail thats sent to you, and the USPS can encrypt the images that are sent to you...

But you still need to receive things that are physical (like credit cards)

There used to be mail services that people could have their mail routed to...

But yeah, we still need mail - but we need "Mail+Pro" Subscription - And now the "You have mail" guy can be replaced by AI...