Walter Ikamba

Finance and Policy grad

[Henry Kabogo 💧 ❄ 🇰🇪 @Kabogo_Henry]

This analysis has premium stitching..it will leave you in stitches..Engineer Joakim Simiyu..🫡🫡

[Random World Ψ @randomworldke]

I knew one sharp Kenyan was gonna connect the dots. I always say Mulima people was not hating on Nyoro. They have this information before hand. His mentor mentored him to be like him 😸😹

[Kirill @kirillk_web3]

A Chinese developer just explained the shift from Loop Engineering to Graph Engineering better than anyone.

most people are still building agents the way that's about to be obsolete.

> why single-agent loops break and go "goal blind"
> the 4 parts of a graph: nodes, edges, state, policy
> 3 topologies that run everything: diamond, supervisor, pipeline
> Anthropic's 5 official workflow patterns

the punchline: it's not how many agents you run. it's the determinism you build with verifiers, code fallbacks, and reality anchors.

I broke the same architecture down with Kimi K3. Full A-Z guide below.

[BrianRoberts221 @BrianRoberts221]

@WesRoth Closer and closer to the Prime Radiant and psychohistory…

[Wes Roth @WesRoth]

Researchers in China have built an AI-powered social simulation that can model more than ONE BILLION human-like agents at once.

The system is called Light Society.

Each simulated person can carry demographic traits, beliefs, goals, memories, social connections and changing internal states, allowing researchers to study how large populations behave and influence one another.

The team built agents from real demographic profiles in the World Values Survey and tested the system on problems including:

→ trust and reciprocity
→ opinion propagation
→ social influence
→ polarization
→ collective behavior

In one experiment, researchers created a network of 1 billion agents and simulated how opinions spread between influencers and the rest of the population over 100 rounds.

Previous LLM-based social simulations generally topped out around 10 million agents.

Light Society pushes that scale roughly 100× further.
But there is an important technical caveat:

This isn't one billion copies of Gemini individually thinking in real time.

To make the experiment computationally possible, the researchers used Gemini 2.0 Flash to generate hundreds of thousands of example interactions, trained smaller surrogate models on those behaviors, and precomputed a lookup table covering 900 million possible interaction combinations.

That optimization is arguably what makes the work so interesting.

If simulations like this become accurate enough, researchers could eventually test policy changes, information campaigns, market behavior, epidemics or social dynamics on synthetic populations approaching the scale of entire countries or the planet.

The researchers themselves warn that the same technology could potentially lower the barrier to designing large-scale information operations.

[𝗩𝗶𝗴𝗶𝗹𝗮𝗻𝘁𝗲 𝗩𝗮𝗻𝗴𝘂𝗮𝗿𝗱 @vinguard254]

Ng'ang'a Muigai once said Senator Sifuna doesn't know the hope that he carries.

[Forecasting Research Institute @Research_FRI]

In his recent blog post, @slatestarcodex wrote that "the AI superforecasters are here."

We run ForecastBench—a leading AI forecasting benchmark and the only one to compare AI forecasters to human superforecasters. So, how close are current AI systems to superforecasters?

For the first time, several AI models are now statistically indistinguishable from superforecasters.

[Rahul @sairahul1]

someone just open-sourced the entire "loop engineering" playbook.

for free. 8.6k stars on github for loop engineering

not prompts.

loops.

everything you need to build autonomous AI systems:

→ scheduling
→ memory & state
→ planning
→ sub-agents
→ verification
→ worktrees
→ MCP
→ stop conditions
→ safety guardrails
→ observability
→ cost tracking

notice something?

none of these are prompts.

they're the systems wrapped around the model.

that's where AI engineering is heading.

we're moving from:

Prompt Engineering → Context Engineering → Harness Engineering → Loop Engineering

soon you'll hear people talking about Graph Engineering too.

because once one loop works...

the next challenge is coordinating hundreds of them.

that's exactly what I explain in:

save this.

I think "loop engineering" is about to become one of the most important concepts in AI.

[Ishika Rawat @Ishh_021]

Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. Stanford taught the entire thing in 1 hour lecture & released it for free.

Bookmark & watch this today before someone takes it down and read this article below.

[Om Patel @om_patel5]

THIS GUY LIVES UNDER SFO'S TAKEOFF PATH SO HE BUILT A CEILING PROJECTOR THAT TRACKS EVERY PLANE FLYING OVER HIS HOUSE IN REAL TIME

he uses a cheap $30 radio receiver to pick up the signals that planes broadcast while flying.

then projects them onto his ceiling in real time

when a jet flies over his house you hear it outside and at the exact same moment a plane glides across his ceiling labeled with the airline, aircraft type, and destination

pure black background so the projector's rectangle disappears and only the aircraft are visible

but he didn't stop at planes

it also draws the real sky behind them. sun, moon, bright stars, constellations, and live satellites including the ISS. all at their true positions for his exact location and time in real time

so he's lying in bed watching the actual night sky projected onto his ceiling with real planes crossing through it as they take off from SFO

there is a huge market for every man alive that runs outside to see the helicopter

vibe coded the whole thing himself with a cheap radio, a projector, and some clever software

[Shubham Saboo @Saboo_Shubham_]

https://t.co/ttTq8aAn1j

[Walter Ikamba @WWleros]

@AliChiapo @FGaitho237 @eastersins Ladies and gentlemen a pure bred idiot
Rarely spotted widely recognizable

[Сarm1ne @carm1nee]

Paul Tudor Jones predicted the 1987 crash, made $100 million, then spent years trying to destroy this footage

you will watch him lose $6 million in one afternoon, sit in his chair and say "total devastation" then make it all back with 100% interest

This documentary will change how you think about risk forever

Bookmark & watch it. Then read the post below - $90 billion from being right just 54% of the time↓

[Val¹ @FormulaValy]

Full Answer

If you can spare four minutes to watch this, I think it really captures who he is. If it doesn’t change your opinion, that’s completely fine just try to watch with an open mind.

[Shanaka Anslem Perera ⚡ @shanaka86]

MICHAEL BURRY JUST WALKED AWAY FROM WALL STREET

Not because he was wrong.

Because being right no longer matters.

His final investor letter contained one sentence that should terrify every fund manager, central banker, and policymaker on Earth:

“My estimation of value in securities is not now, and has not been for some time, in sync with the markets.”

This is not a crash prediction.

This is a death certificate.

THE NUMBERS NOBODY WILL TELL YOU

Passive index funds now control 52% of all U.S. fund assets. Fifteen point four trillion dollars. BlackRock, Vanguard, and State Street collectively manage twenty five trillion. They own dominant stakes in virtually every public company in America.

They analyze nothing.

Every dollar flowing into an S&P 500 ETF automatically allocates thirty five cents to seven stocks. Not because someone studied the balance sheets. Because that is their index weight. The algorithm cannot read an earnings report. The algorithm cannot process overvaluation. The algorithm simply replicates.

THE DEATH SPIRAL

Passive buying increases prices. Higher prices increase index weights. Higher weights attract more passive buying. The feedback loop operates completely independent of whether any business is worth what the market claims.

The marginal buyer of American equities is no longer an analyst with conviction.

It is a target date fund receiving its biweekly paycheck allocation.

This buyer has no opinion.

This buyer will never sell.

THE CONSEQUENCE

Price discovery, the mechanism that has allocated capital for two centuries, has been structurally disabled.

The most successful short seller in modern history looked at these numbers and concluded that fundamental analysis no longer converts to returns.

He did not predict collapse.

He diagnosed something worse.

The market still exists.

The market no longer thinks.​​​​​​​​​​​​​​​​

Read the full story here 👇

https://t.co/ubcya24WC8

[I am Chege @_James041]

How did I fail to watch this article? This was on 5th March 2025.

Okiya Omtatah submitted a petition at the Milimani Law Courts detailing how SHA is a fraudulent scheme.

This clip is only 3 minutes long. It will make you even more angry.

#RutoMustGo #DrainTheSwamp

[Dylan Small @_DylanSmall_]

Huge 3D Printing Breakthrough!

Caltech is printing metal atom by atom. A hydrogel scaffold soaked in metal salts becomes the blueprint, the gel burns away, and oxygen is stripped in a hydrogen-rich chamber, leaving alloys crafted with surgical precision.

They can change metal ratios on the fly. Flip from a copper-heavy mix to a Cu12Ni88 alloy nearly four times stronger, without altering shape or tooling. Every property, from strength to density, is now designed, not guessed.

This is metallurgy reinvented. Also a game changer for Aerospace applications.

https://t.co/Wcr2mNaKgb

[Typical African @Joe__Bassey]

In Papua New Guinea 🇵🇬 this young men created an intriguing musical sound from striking a bamboo pole stick.

[Massimo @Rainmaker1973]

He created a device that lets paralyzed individuals control phones, tablets, and computers with only their tongue.

Created by MIT-trained engineer Tomás Vega, the device sits on the roof of the mouth and works like a wireless trackpad.

[Walter Ikamba @WWleros]

@Dr_AustinOmondi nkt