saejox

i believe a harness is a personalized thing. everyone should develop their own. doesn't take long to have a working prototype, after that you can extend it as you use it.

it is very hard to run 150k token context with these kind of setups. even if you do, all of us are used to running multiple agents at once.

i hope ram prices will come down a little.

turn this in to a puzzle game. i am sure there are some enjoying ikea assembly.

Consistently buggy you mean. 1 in 10 tasks get stuck in reasoning loops or error out.

No human knows the answer to that question or to any any other technical question. that will be the world we live in

i asked copilot it says runtime interception and translation

Actual question is this:

As my weekly/hourly quota is nearing its end, does Anthropic start to serve me a worse model?

Am i personally being throttled? testing against API is meaningless.

i can buy a house with that amount of money. it used to be car money.

This is a project i wanted to implement for a long time. It regularly benchmarks cloud hosted models with private benchmarks. Not just openai & anthropic, popular openrouter models too.

Tests their intelligence, not their diligence.

Sadly i cant think of a way to monetize the service. Also if it ever gets famous enough labs would try to game the system, it would be cat&mouse game that i am not willing to waste time on without any monetary gain.

Not even close to astra. Astra is something else. It is expensive, but uses way fewer tokens do my tasks.

xAI missed its chance, Ball is on Anthropic's court.

Work days on case study only to be hacked and get even poorer.

To develop a smart ai system for my 2d roguelike platformer? game has way too many moving system for classic state-machine ai + i cant spare the time to develop it. its low latency entices me.

Needs to support tcp and udp

it is a feature called eventual consistency.

so nothing will be %100 correct, and we will always live in the past.

like downloading exe files from the web. but we should invent something to catch bad actors putting bad stuff in them

nice. we should call it web, like spiders' web all connected.

idea has merit.

ai slop is an instant turn off for me. i would never be convinced with such low effort website.

Could also have been some H200 selling inference on openrouter with stolen electricity.

Art for art's sake, did not work for me.

I was planning to earn some money from my second Steam game. Than AI happened.

I did not gave up on because competition is now higher. I gave up because programming was fun for me, been at for 20 years. But right i always know in part of my brain 'this 1000 lines of code could have been just a prompt'.

Joy is not there for me anymore. I hope to get it back someday.

my handwritten very detailed 10 lines comment block is %87 robot. and i use broken english with wrong grammer all the time.

you are not the only one, Pangram and GPTZero also think it is written by a bot.

ai detection may be an unsolvable issue.

they add feature after feature. but never a good drive for the player. 'why should i go to this planet?'

i have an idea for it,

Let users populate the universe with custom quests. Scriptable quests with dialog and multipart structure. You need a custom script language and some limitations on it. Example quests:

- "Bandits have occupied our town", but actually bandits are tax collectors. You instead may help them document the goods.

- "Our star is dying", instead a huge starhip is recharging and refusing to move without some condition.

also add a rumor system to lead players to quests.

you can use my idea Hello Games, you have my permission.

another good target for ai rewrite: getting out of serverless cost hell.

it used to be a gigantic endeavor. now a few days work.

make it 4x40 raid-0 ssds to achieve 40 tps.

or 40 macbooks with each 4 ssd. to get 40 tps.

Meta as a company is too successful. It's influence on the western world is undeniably extensive.

Looking at the this picture & share holder value, Zuck is a successful CEO. I doubt any Meta insider would like Zuck to step-down.

I for one, would like Meta to crash & burn. People need to be free of its influence.

i agree, skills downloaded from the internet are all snake oil.

creating your own skills however good for both reducing the token usage & increasing reliability. those damn llms are not deterministic, asking same thing twice produces 2 different results.

i agree with good pr and common good. however,

challenges:

articulated robots are expensive (many would steal them) dust and water resistance with moving parts

Someone will create a self replicating robot factory soon.

All robots designed by robots.

Likely a crazy youtuber first

there are two sides of this argument.

only way we are able to eat affordable food is because some diesel truck is transporting it.

so maybe we must subsidize diesel fuel, to decrease grocery prices?

it was at least another thing worth flagging

AI written article criticizes AI written articles.

Even the title is AI written, show some effort people.

Performance or quailty of the llm is not the draw for me.

I read those blog posts to remind of wealth gap i have with the average hackernews user.

Thank you for the encouragement, i will work harder to reach your level.