UX Designer. Google Design Sprint Master. Creating brands, websites, and digital products for visionary minds and industry innovators.
Planet Earth
Joined July 2007
- Tweets5.9K
- Following2.4K
- Followers1.2K
- Likes11.4K
Esto es oro puro.
Todo lo demás es que te venden es humo.
ESTO ES ORO PURO
SpaceXAI acaba de publicar un manual de 3 páginas para convertir Grok Bot en un equipo de IA que trabaja 24/7
El cambio es simple:
En vez de darle tareas a un solo Bot una por una, construyes un Chief + equipos de especialistas que se encargan de workflows completos
Aquí tienes el playbook de Grok Bot en 7 pasos:
Paso 1: modelo mental
Un único pc en la nube, múltiples pantallas de Bots, archivos compartidos, sesiones del navegador, credenciales y /workspace funcionando por debajo
Paso 2: orden de configuración
Crea primero el Chief.
No hace el trabajo, enruta → delega → supervisa los handoffs → recopila resultados → escala cuando es necesario
Paso 3: bus interno
Deja de pasar contexto importante por el chat.
La investigación, borradores, evidencias, decisiones y handoffs viven dentro de /workspace, para que cualquier Bot pueda continuar exactamente donde lo dejó otro
Paso 4: equipos de especialistas
Divide contenido, inteligencia, operaciones, investigación y coding entre distintos Bots mientras un único Chief coordina todo
Paso 5: automatización
Cada workflow serio tiene 3 puertas: source gate → evidence gate → action gate.
Los Bots siguen trabajando hasta que realmente necesitan aprobación humana
Paso 6: capa de confianza
Deja que los Bots investiguen, redacten, resuman, reconcilien, verifiquen y preparen acciones de forma autónoma.
Los humanos solo intervienen para dinero, publicaciones, eliminaciones, firmas y acciones externas irreversibles
Paso 7: lecciones caras
Hacer polling constantemente quema la cuota, sincronizarlo todo desperdicia uso, los malos reintentos pueden duplicar acciones, los sistemas internos no documentados terminan rompiéndose y añadir demasiados Bots puede hacer que todo vaya más lento
Resultado:
Grok Bot deja de ser un simple chatbot y se convierte en un equipo de IA 24/7:
→ Cada workflow tiene un responsable
→ Cada handoff deja evidencia
→ Y los humanos solo intervienen cuando hace falta criterio o aprobación
Lee esto antes de crear otro Grok Bot
El manual reúne las 18 reglas y todo lo necesario para montar un sistema multiagente: arquitectura, handoffs, aprobaciones y automatización
Guárdalo y construye tu propio equipo de Grok Bots con el artículo de abajo👇
Leandro Henflen - Product Design retweeted
I built a Gamma skill that writes enterprise pitch decks.
In 3 minutes it does what my $15K designer took 10 days for.
→ No more $2K retainer per designer brief
→ No more 48h turnarounds and 4 rounds of revisions
→ No more 10-day latency between pitch intent and deck delivery
→ No more losing deals because the clock ran out
→ No more generic templates VCs have seen 50 times
Just one prompt → custom 30-slide enterprise pitch deck in 3 minutes.
Here's how it works:
→ ChatGPT Memory Integration (pulls 6 months of client calls, positioning, pricing logic)
→ Skill Architecture (hits Gamma API directly, no manual slide building)
→ Narrative Generator (custom story per prospect, not recycled templates)
→ Case Study Pull (surfaces real wins from past call transcripts)
→ Brand Voice Layer (decks sound like you wrote them, not an AI)
→ 12-Prompt Playbook (launch, refine, repurpose, repitch)
Built on ChatGPT Connector infrastructure.
Runs on your own business memory, not generic prompts.
3-minute deck generation. 0 designer dependencies.
Results from 11 days of deployment:
- $45K retainer closed on first deck sent
- $180K total pipeline closed
- 3 more prospects decked in the same week
- 0 revisions across all 4 decks
The job isn't making better decks. It's killing the 10-day gap between intent and close.
Reply "PITCH" + repost, I'll DM the skill + 12-prompt playbook.
Flow
MakeUGC = Zero-Cost Creative Scale
Fully realistic "Product-in-hand" ads powered by AI agents. No cameras, no creators, no shipping products.
- Vibe: Pinterest-style UGC aesthetics
- Speed: Minutes from prompt to export
- Platform: Optimized for TikTok, Reels, & Paid Ads
- Motion: Cinematic movement (Veo) & realistic hands (Kling)
One AI engine that generates and scales your brand content 24/7.
Comment + RT “FLOW” and I’ll send you the link.
(Must be following)
Leandro Henflen - Product Design retweeted
Just built a $5,000 gym software in 48 hours and it’s borderline unfair 🤯
It books leads, follows up with members, and brings back lost signups on autopilot
Just a simple AI system sitting between Instagram DMs and gym revenue
Full stack:
* Claude → conversation logic + scripts
* Make.com → automation flows
* Google Sheets → member + lead database
* WhatsApp API → instant follow-ups
* ChatGPT → offers, onboarding, retention messages
What it actually does:
→ New lead comes in from IG or ads
→ AI replies like a human PT within 5 seconds
→ It qualifies the person (goal, budget, timing)
→ Sends them a trial offer automatically
→ If they ghost, it follows up for 7 days without sounding desperate
→ Existing members get reactivation nudges when they stop showing up
Build time: ~1–2 days
Value: replaces a front-desk + sales follow-up system gyms usually overpay $5K+ for
Most gyms don’t lose members because of price.
They lose them because nobody follows up.
This system fixes that silently in the background.
No new staff.
No extra workload.
No “come back” awkward calls.
Just consistent revenue recovery on autopilot.
If you want the exact system breakdown + prompts:
1. Comment "AI"
2. Repost
I’ll send you the full build.
(must be following so I can DM)
Leandro Henflen - Product Design retweeted
Claude = 550 videos/day
Fully realistic UGC ads — cinematic lighting, natural human motion, clean pacing — powered by AI agents.
- UGC cost: $0
- Production time: minutes
- Scale: instant
One AI engine that creates, tests, and scales short-form ads automatically — nonstop.
Comment + RT “V2” and I’ll DM you the full workflow.
(Must be following)
Leandro Henflen - Product Design retweeted
We built our launch video in Claude Code using HyperFrames.
Now it's yours.
Open source, agent-native framework. HTML to MP4.
$ npx skills add heygen-com/hyperframes
RT + Comment "HyperFrames" to get the full source code of this launch video (must follow)
Leandro Henflen - Product Design retweeted
An easter egg skill we hid in hyperframes:
/website-to-hyperframes
- create DESIGN.md
- screenshot the page - download assets
- build logo animations + more
we hope to support the launch of anyone's businesses
RT + comment "Website" for codebase access (must follow)
Leandro Henflen - Product Design retweeted
i genuinely should be gatekeeping this but f*ck it:
FULL breakdown of the exact manus ai prompts i use to build high-converting ecom landing pages in MINUTES
for 24h, i'm sending it to EVERYONE who likes + comments "PAGES"
(must be following + RT for priority access)
Leandro Henflen - Product Design retweeted
Cleared ~$182K running this exact setup recently…
Most people in ecom aren’t stuck because ads don’t work
they’re stuck because they overbuild everything before testing
this is way simpler than you think
- pick a product already converting
- generate multiple AI UGC angles fast
- test hooks, not “campaign structures”
- scale only what actually gets attention
no long production cycles
no waiting on creators
no guessing what might work
just speed + volume + iteration
same system, different creatives, over and over
rt + comment “ugcflow” and i’ll send the full breakdown
(follow for dm)
This is getting better by the minute
Leandro Henflen - Product Design retweeted
I built 7 Claude Skills that replaced 6+ hours of weekly marketing work.
Now I’m giving them all away.
The Founder Skills Vault:
→ Viral Content Generator
→ Content Pillar Generator
→ LinkedIn Content Analyzer
→ Lead Magnet Idea Generator
→ LinkedIn-to-X Converter
→ Warm DM Strategist
→ Lead Qualifier
This isn’t a prompt pack.
These are production-grade tools that run inside Claude.
It’s the same system I used to grow 7,000+ followers in 3 months.
Want it?
→ Like + RT
→ Comment “Send”
Leandro Henflen - Product Design retweeted
¿Cómo era el tema de cazar en un zoológico? Por fin la gente se está dando cuenta de la estafa de los precios dinámicos, como que la gente ahora usa MercadoLibre como comparación de precios o referencia y no tiene otra forma de usar referencias, por ejemplo, un local físico en la calle. Ahora te ponen precios con el algoritmo o IA, que saben tus movimientos y gastos; por ende, a algunos les aparece más plata y a otros menos. Me gustaría decirles un truco para engañar momentáneamente al algoritmo, pero los putos lo leen y lo parchean. Me viene funcionando más o menos...