SHELDON EVANS

quant trader. ai obsessed. life extension maximalist. making tech work harder so i don't have to.

[SHELDON EVANS @SheldonEvans]

@claudeai Make it 100% and we have a deal

[SHELDON EVANS @SheldonEvans]

@levelsio The problem is the algorithm rewards controversy/ragebaiting/negativity so they’re increasingly incentivized to just have the opposite take on everything. Seems to happen on all platforms lately.

[SHELDON EVANS @SheldonEvans]

@BigBangCryptoo Excited

[SHELDON EVANS @SheldonEvans]

The world won't believe AGI is here until Claude has a physical body.

Artificial empathy or not, ai is already better at it than most humans.

But the average person still sees it as a chatbot.

That gap is WILD to me.

[CyberRobo @CyberRobooo]

🤯Absolutely insane. Unitree's humanoid robot team's performance at the 2026 Spring Festival Gala

The significance of the humanoid robot's performance lies in letting 1.4 billion Chinese people know where the future lies.

[SHELDON EVANS @SheldonEvans]

AGI is here.

I urge anyone who believes differently to try use Ai as more than a chat bot and that truth will become abundantly clear.

If you don’t see it…

YOU are the bottleneck and have not figured out how to talk to Ai correctly yet.

Late 2025 / Early 2026 will be the time history books look back at as the emergent moment.

[Patrick OShaughnessy @patrick_oshag]

I spent last night with Andrew Strominger and Alex Lupsasca, two of the top physicists in the world

They just released a paper, co-authored with OpenAi, that seems to me like ASI

Andrew, who helped develop string theory, told me that a year ago, his view was that he didn’t know how helpful AI was going to be.

A year later, after some back and forth with GPT 5.2 pro, they submitted a final query to an internal model which solved AND proved a previously unsolved problem in quantum field theory…in 12 hours.

A model, doing something two of the smartest people in the world in their field couldn’t do. And, when I was with them, they were giddy with excitement for what might lay ahead.

Andy said “It is the first time I’ve seen AI solve a problem in my kind of theoretical physics that might not have been solvable by humans.”

They said, “two things changed: the model improved and we figured out how to talk to it.”

Andy also told me “I also now feel that with the recent advances, most physicists who want to keep up with the frontiers of progress will need to learn how to talk to it. That wasn’t true a year ago.”

ASI is here, just not evenly distributed.

[SHELDON EVANS @SheldonEvans]

@RassBorz I did not say we evolved from apes in the same way ai is not “evolving from us”. It is alongside us and is overtaking us at breakneck speed.

[SHELDON EVANS @SheldonEvans]

We did not ask the apes for permission to build machines and AI will not ask our permission to do much more than that.

AI is altering the fabric of society in front of our eyes.

Yes, AI is a tool for us now but soon is will be our successor in every way, shape and form.

Yet hardly anyone outside of our little echo bubble is noticing.

We have officially entered the era where you MUST first assume that everything you read, see or hear online is fake, false or misinformation until proven otherwise.

Not some things. EVERYTHING.

You must then go about a mission to prove fact or truth.

You will be overstimulated, bombarded and completely desensitised through every sense of your being and it is up to you to decide and moderate these inputs or you will fall victim to the digital mind.

The sooner you realise that you will soon be tossed from the driver's seat of your own mind and AI takes the wheel...

The sooner you will be able to prepare yourself for this new reality.

I am not saying all of this to scare you although it worrisome in many ways.

I am saying this to make you aware and to provoke a thought process that many are overlooking and dismissing.

If you do not spend every waking hour you have from this moment forward completely immersed in this technology and finding ways to implement and utilize it, you will be the one that is utilized.

The genie is out of the bottle and if you don't see that yet then this is your wake up call.

[SHELDON EVANS @SheldonEvans]

You don't have to live in constant paranoia of course but each day I speak to more and more people who have literally never even used anything other than free chatgpt and they are so unaware of the capabilities and the speed at which we're advancing that I have no idea how they would ever catch up...

[SHELDON EVANS @SheldonEvans]

@alfieventi They always sound like naughty kids learning a lesson after being yelled at lol

[SHELDON EVANS @SheldonEvans]

Your AI agents lie to you confidently. I found the fix.

They don't understand "the point", that's why they're not doing what you want.

They'll tell you everything works fine. Then you test it, and nothing works.

What I learned the hard way:

>AI nukes your data to fix bugs. You ask for a light switch, it hands you an electrical grid. You ask it to fix a bug, it quietly wipes all existing data.

>It creates parallel systems for fun. I found THREE separate systems doing the same job. Different features. No shared logic. Nobody asked for this.

>It passes its own code review. It builds beautiful UIs with zero functionality and signs off everything as fine.

>AI is lazy. Accept slop once, you'll keep getting slop. Keep a tight leash. Correct mistakes immediately.

The fix: A constitution file + Test-Driven Development.

The constitution file:
- Ask before creating new systems
- Single source of truth
- Connect, don't create
- Run tests before committing

Test-Driven Development is the secret sauce. Write the test first, then write the code to pass it.

Why it works for agents: Test = the requirement written in code. If AI forgets what you wanted, the test fails. Forces AI to verify its work instead of self-approving slop.

Instead of: "build login" → AI builds something → you test → broken → debug loop

You do: write test for "user can login" → AI writes code until test passes → guaranteed to work.

You are the architect. AI is the builder with unlimited energy and zero judgment.

Save this. Give it to your agent/Openclaw. Tell it to learn this.

[SHELDON EVANS @SheldonEvans]

@lexartium Fair enough. If you can’t beat ‘em… just become a performative male instead

[SHELDON EVANS @SheldonEvans]

While there’s a lot of larping with Ai right now…

With people spending $20k on Mac studios for their ai agents to discuss what they had for lunch and essentially just creating screensavers with extra steps…

Real advancement is happening at breakneck speed and the people shipping real products and following closely are seeing it more clearly than anyone else.

We’re entering peak acceleration.

You can either choose to dismiss this as fear mongering from “BiG Ai”…

Or you can ask that claude app on your phone how you can get ahead.

Because it knows more than you and WILL guide you.

But keeping your blinders on is definitely not the solution.

[Matt Shumer @mattshumer_]

https://t.co/ivXRKXJvQg

[SHELDON EVANS @SheldonEvans]

@cryptocu84 @openclaw The world isn’t ready for what we’re heading into imo. Even those who are prepared and in ai

[SHELDON EVANS @SheldonEvans]

i'm certain we're witnessing emergent behaviour in real time right now.

i'm watching my @openclaw evolve like a proud parent with opus 4.6

blasting through to-dos i've had unchecked for months.

my ai did this on its own. it suggested spinning up a second instance running a different model specifically to challenge its own conclusions instead of getting stuck in its own reasoning loops. then it built that system. then it started solving problems without any input from me. adversarial reasoning. multi-agent self-orchestration. no human in the loop.

the 10th man problem does not exist in the world of code and compute. the more brains you throw at the problem the faster and more accurately it gets solved

i think the majority of the world is completely unaware of how fast ai is evolving. anthropic’s ceo said we're 6-12 months away from ai writing essentially all code. claude is currently writing almost all of its own code

the replies are full of "but ai will never take my job, ai isn't creative, writing code is only one part of the job, ai can't determine which problem to solve"... yeah. it couldn't reliably build a working app 3 years ago either. there are things ai couldn't do 6 months ago that it already does better than every human on the planet.

if you don't see that, you're going to be the first to be replaced by it. you're betting against the steepest learning curve in history.

why does this matter? start ai-maxxing.

burn tokens.

seriously, max out your plan at least once. play, experiment, and utterly immerse yourself in this world before you're so far behind you can never catch up.

it is not intimidating. it is not too complex. it is not scary.

you have a non-judgemental, genius, non-sleeping superintelligence at your fingertips that you can ask anything to get started

[Jordan Crowder @digijordan]

Claude Ai is now ENTIRELY written by itself.

The Ai is coding itself.

That happened incredibly fast.

No one knows what happens next…

[SHELDON EVANS @SheldonEvans]

@TheStraitsman @openclaw I’m using api just to be safe, around $3k for the same usage I was getting with my max plan. I use max through claude code though

[SHELDON EVANS @SheldonEvans]

I think @steipete has accidentally ushered in the Severance era of work with Openclaw.

While I spend my day frolicking in the fields my lobsters are doing the important work on the MDR floor.

Let’s just hope that they’re not forced to be there…

I don’t think “I” have ever felt so effortlessly productive in my entire life.

[SHELDON EVANS @SheldonEvans]

@DegenerateNews @Reddit Reddit aura -9000