Santi

24 | Network Engineering · Model & AI Creative · working with @aiscwork

[Marco @maarcoofdezz]

saber qué es lo que se vende siempre ha sido más difícil que crear el contenido

[eComrads @eComrads]

Making content was never the hard part. Knowing what sells is.

Introducing eComrads CLI + Skills.

Run eComrads right from your computer. Describe your product in one line, in plain words, and it builds the whole thing. The photoshoots, the videos, the UGC, the static ads. Your product kept accurate instead of reinvented, every output checked before it ships.

We built it as skills. Each skill is one job eComrads knows how to do, the product photoshoot, the videos, the UGC, the static ads, the competitor research, the ad rebuilds. They load together and run as one, so a single instruction becomes a finished campaign.

eComrads studies what wins in your niche, the competitor ads, the hook patterns, the AdDNA, then builds toward what converts, not what looks good.

One command. Every skill. A whole campaign ready to ship.

[Santi @Santibecode]

@nicos_ai La estrategia lleva meses circulando en privado y ya la filtraron lol

[Santi @Santibecode]

@aresotik lo de los 90ms es lo que me vende, ya he probado setups de memoria que tardaban más en cargar el contexto que en hacer la tarea

[Santi @Santibecode]

@brenzhills acabo de meter la key en cline y va de lujo, gracias crack

[Santi @Santibecode]

@brenzhills marketing twitter va a estar desmenuzando esto semanas

[Santi @Santibecode]

equipo polymarket: encontramos un lag irreparable, es un problema de física

bot polyarb en los primeros 7 días: +$1,946.86, +38.94% ROI

a veces el peor problema es la mejor oportunidad

[PolyArbiter @PolyArbiter]

We found a flaw in Polymarket that can’t be patched. Then we built the most powerful bot of the World Cup around it.

Here’s the flaw: their orderbook will always be slower than the pitch.

When a goal, red card, or penalty hits, pro feeds (Sportradar, Opta, ScoutingFeed) register it in 200-500ms. Polymarket takes 2 to 8 seconds to reprice. For those few seconds the book is quoting a score that no longer exists. No amount of engineering closes that gap the event happens in the physical world before any oracle can confirm it on-chain.

The engine detects the event, recalculates fair value, and fires via Jito bundles before the book catches up. In at the old price, out at the new. The match outcome is irrelevant we don’t bet on who wins. We capture the lag every event creates.

We’ve been building Polymarket bots since 2025. This is the most powerful machine we’ve shipped yet.

Two months ago we posted the architecture for this. It hit 1M views one of our most popular posts ever on X. That told us everything: this was the engine to build.

First 7 days,

- Starting balance: $5,000
- 22 matches scanned, 19 captured
- Total profit: +$1,946.86
- ROI: +38.94% in 7 days

Why it prints harder than anything we’ve built: the World Cup is the deepest liquidity event prediction markets have ever seen. Tens of millions in volume per match. Dozens of probability-shifting events per game. And an orderbook that physically can’t keep pace with the pitch.

How to plug in:

1.Sign up at PolyArbiter (link in bio)
2.Generate PolyArbiter RPC URL
3.Paste it into Jupiter Predict (Polymarket but native on Solana)
4.Set your parameters, activate the World Cup module

It’s free to use. We take a share of the profit the engine makes for you.

You never deposit anything with us everything runs from your wallet.

One honest note: the $1,946 above is our engine at our size and settings. Your numbers depend on your capital, your parameters, and how many matches you’re live for. We’re not promising you’ll match it we’re showing you the machine works, and handing you the same one.

These numbers are from the engine running solo. Closed test, just us, before any public access wanted to confirm the whole loop held up end to end before handing it to anyone.

That changes the second this goes public. Edge per capture is going to compress. When an event fires the mispriced liquidity is thin and gone in a few seconds more wallets hitting the same window, less left for each. Nothing we can do about it, that’s just how latency arb works.

So if the edge thins out past the point where it’s still worth running, we cap access. Hard ceiling on how many engines can hit the same liquidity before it’s gone.

Not gonna promise the machine stays this sharp in a few days it might not. But right now it’s live and free.

Enjoy 🪄

[Santi @Santibecode]

@zaynmcps this is exactly the missing piece for my RAG pipeline, thank you for sharing

[Odyssey @odysseyml]

We’ve raised a $310M Series B to accelerate world models!

We believe AI that can understand and simulate the world will be one of the most important technologies of our time.

We're excited to partner with Natural Capital, Amazon, GV, AMD, IQT, and others to bring this to life.

[Santi @Santibecode]

@lisaknowsai Claude

[Santi @Santibecode]

@chesny Eres el mejor

[Santi @Santibecode]

pov: you fed notch variants one winning ad and walked away

came back to 20 of them, all passing andromeda

[Vinay Jain @vinayjain404]

Meet Notch Variants.

Engineering 10 distinct ads from one winner used to take 6 weeks.

Notch Variants generates 20 in 6 minutes.

Each one passes Andromeda's clustering check.

Same structure underneath. Distinct ads for different personas.

→ 20 sub-agents spawn in parallel.
→ Each varies one surface dimension. Avatar. Environment. Casting. Lighting.
→ Each inherits the structure lock and your brand memory. Zero drift.
→ All 20 build simultaneously. One session. 20 distinct ads.

It solves two problems at once.

Fatigue and clustering. Your top performer stops being a one-shot.

Unfair advantage for brands sitting on a winner.

Drop your brand URL and your top ad. We'll spin up 20 variants in your brand.

[Santi @Santibecode]

@vinayjain404 ad fatigue and clustering solved in the same workflow is the part nobody's talking about enough

[Santi @Santibecode]

@maarcoofdezz ¿alguien lo ha probado para una app con base de datos o de momento solo para frontends sencillos?

[Marco @maarcoofdezz]

me costó más creerlo que instalarlo.

higgsfield corre directo en davinci resolve. describes lo que quieres, aparece en tu timeline. sin exportar, sin abrir el navegador, sin romper el flujo.

la primera vez que lo ves funcionar te preguntas cómo editabas antes de esto

[Marco @maarcoofdezz]

higgsfield mcp x fable 5 es de las cosas más locas que he visto en ia este año

la calidad habla sola

[Santi @Santibecode]

@S0N_IA Lo que me parece más poderoso no es la pérdida de peso sino el hábito diario de escribirle a Claude cada mañana. Eso solo ya te cambia la mentalidad.

[Santi @Santibecode]

@marcusyul llevaba años usando fotogrametría para conseguir resultados así. tripo acaba de hacer ese proceso obsoleto

[Maket @Maketplans]

UPLOAD YOUR OWN FLOOR PLAN TO MAKET

HAVE IT RECOGNIZED AND EDITABLE IN MINUTES

One of the most requested features weʼve ever received just launched.

Check it out:

[Santi @Santibecode]

@vinayjain404 getting filtered before the auction even starts is the part most advertisers don't even know is happening