kokis

Always go too far,
because that's where you'll find the truth.

Albert Camus

[kokis @damigos_george]

ΕΠΑΛ Γυθείου. Δύο μαθητές μου απο φτωχές οικογένειες, που μένουν σε χωριά που δεν τα έχει ούτε ο χάρτης, μπήκαν ο ένας Ηλεκτρολόγος Μηχανικός στο ΕΜΠ και ο άλλος στο Πληροφορικής και Τηλεπικοινωνιών στο ΕΚΠΑ. Είμαι πολύ περήφανος για σας μάγκες...

[しらき@パワポ図解 @kumiko_shiraki]

【NotebookLM】オールドメディア風のスライドが作れるYAML…プロンプトはリプ欄へ↓

[しらき@パワポ図解 @kumiko_shiraki]

177:
Design Style:
Name: "Cinema Ticket"
Concept: >
映画館の入場券、35mmフィルム、映写機、クラッパーボード、
レトロな映画祭プログラムから着想を得たシネマ・チケット・エディトリアル。
各スライドを映画のチケット、フィルムフレーム、上映案内のように構成する。
Canvas: {Ratio: "16:9"}
Mood: [Cinematic, Vintage, Editorial, Nostalgic, Dramatic, Structured, Sophisticated]
Background:
Primary: "#F3E9DC"
Secondary: "#2B2A27"
Support: "#777B6D"
Texture: [生成りの厚紙, 微細な紙繊維, かすかな印刷ムラ, マットな質感]
Color Palette:
Base: ["#F3E9DC", "#2B2A27", "#777B6D"]
Accent: ["#D7A95E", "#21B9A8"]
Rules: [生成りとチャコールを中心に構成, オリーブグレーを背景に使用, ゴールドは引用枠, ターコイズは小さなコーナーアクセントに限定]
Typography:
Heading: {Font: "Condensed Cinema Sans", Weight: "800-900", Style: [Uppercase, Tall, Bold, Poster-like], Color: "#2B2A27"}
Body: {Font: "Clean Sans Serif", Weight: "400-500", LineHeight: "1.45"}
Labels: {Font: "Condensed Sans Serif", Weight: "700", Style: [Uppercase, Compact]}
Layout:
Grid: [映画フィルムのコマを模したグリッド, 太い黒枠, 上下にパーフォレーション, 左右分割, 広い写真領域]
Composition: [チケット型表紙, クラッパーボード型導入, フィルム帯の目次, 人物写真を並べるチーム紹介, 装飾チケット型引用, 写真と説明の左右分割, データとフィルム写真の組み合わせ]
Cinema Elements: [35mmフィルムの穴, チケットの半円ノッチ, バーコード, クラッパーボード, 映写機, フィルムリール, 小さな映画アイコン]
Images:
Style: [Cinematic Still, Film Photography, Behind the Scenes, Portrait, Projector and Reel]
Treatment: [低彩度, 暖色寄り, 粒子感, 長方形トリミング, フィルムフレーム内に配置]
Decorative Elements: [フィルムストリップ, チケット枠, 太い黒線, 星, バーコード, 三角形の再生マーク, コーナーの小さな色面]
Effects: [Film Grain, Paper Texture, Ink Wear, Flat Print, Subtle Shadow]
Rules: [すべてを同一映画祭のプログラムとして統一, フィルム穴とチケット形状を主要モチーフにする, 色数を抑える, 写真は映画的かつ低彩度, 角丸UIや現代的SaaS表現は避ける]

[Beginnersblog @beginnersblog1]

6 Genre-specific Prompt Styles in 2026

— Horror: Candlelit, low-key lighting, 16mm grain, slow push-in, crushed blacks. In the style of Robert Eggers.

— Sci-fi: Brutalist scale, volumetric god rays, IMAX 70mm, cool cyan/magenta grade. In the style of Hoyte van Hoytema.

— Music Video: Neon alley, CineStill 800T halation, anamorphic flares, tracking shot. In the style of Hiro Murai.

— Indie: Handheld, natural window light, muted palette, diegetic sound only. In the style of Lynne Ramsay.

— Documentary: Locked off, single soft key light, no fill, naturalistic grade, room tone.

The lighting, lens, and film stock references aren't decoration. They're the actual keys the model reads to unlock a genre's visual DNA.

Master the vocabulary, control the aesthetic.
.
.
.

[SilenceÇaPrompt @SilenceCaPrompt]

🚨 DERNIÈRE MINUTE : Claude peut maintenant te créer un site web de luxe en seulement 30 minutes.

Et totalement GRATUITEMENT.

Voici 8 prompts pour créer le site web de tes rêves 👇🏽👇🏽

[Libros y escritores @Librosyescritor]

DOCUMENTAL COMPLETO | Este es el filósofo al que llamaron hereje y de quien Albert Einstein dijo: «Creo en el Dios de Spinoza»

Baruch Spinoza fue condenado al silencio. Excomulgado por su propia comunidad. Tachado de blasfemo y perseguido por sus ideas. Pero no se extinguió. Se volvió eterno.

Descubre la historia del hombre que redefinió a Dios, que desafió la ortodoxia religiosa y cambió la historia del pensamiento sin levantar la voz.

[Akanksha @Akankshaku46881]

Stop prompting Claude like everyone else.

Clone your thinking, writing style, and context into Claude with this 7-step workflow.

You'll get: ✅ More personalized responses
✅ Consistent output every time
✅ Less time repeating yourself

Save this thread for later. 🔖

[Gadgetify @Gdgtify]

Another prompt for mind-bending science facts. It's not hard coded, so you can guide it

2x2 grid, 16:9, 4 mind-bending science facts input = [subject / fact_or_stat / reference_medium / data_mode] run science_fact_visualization_extraction: subject_dna := infer what subject is, its scale, material, shape, and recognizable visual identity fact_type := infer whether the fact is density, size, mass, speed, heat, pressure, distance, probability, lifespan, energy, scarcity, cost, or another measurable category measured_value := use provided value or verified value; if illustrative, clearly mark as illustrative reference_comparison := infer what the value is being compared against visual_metaphor := infer the most intuitive physical demonstration of the fact environment := infer scene that makes the comparison visible physical_interaction := infer floating, sinking, stretching, melting, compressing, orbiting, stacking, shrinking, overflowing, balancing, glowing, freezing, or other behavior callout_text := generate concise explanatory label leader_line := connect label to the exact visual evidence camera := infer cinematic educational framing lighting := infer clean, beautiful, credible science-documentary lighting render: a cinematic visual explanation where the fact is embodied by the subject’s interaction with the reference medium. forbidden: hard-coded planet, hard-coded water, hard-coded density value, hard-coded text box, unsupported physics, fake precision without data, purely decorative infographic text.

[しらき@パワポ図解 @kumiko_shiraki]

【NotebookLM】線画スライドを作りたいときのYAML…プロンプトはリプ欄へ↓

[Dilum Sanjaya @DilumSanjaya]

Fun interactive science app ideas | Part 8

Built an app to explore details about different pollinators

Code
Gemini 3.1 Pro

[しらき@パワポ図解 @kumiko_shiraki]

【NotebookLM】ゆるキャラを作りたい時のYAML…プロンプトはリプ欄へ↓

[Francisco García @Fran_actua]

LLEVAS AÑOS USANDO GOOGLE DRIVE PARA GUARDAR ARCHIVOS.

Pero Google Drive dejó de ser un almacén en 2026.

Ahora tiene IA que busca dentro de tus documentos, escanea textos con la cámara, genera presentaciones solas y responde preguntas sobre tus archivos.

12 funciones que el 95% no ha activado nunca 🧵👇

🚨 Guárdalo.

Tu Drive va a cambiar por completo.

[Ben Pierron @Ben_escrito]

MIT acaba de publicar diez cursos gratuitos de IA para líderes.

Una hora con ellos puede marcar la diferencia entre aprobar iniciativas de IA y simplemente firmarlas sin entenderlas.

La mayoría de los líderes asume que la alfabetización en IA es un asunto de ingeniería.

No lo es. Es un asunto de toma de decisiones.

No necesitas saber programar.

Necesitas entender una capa más profunda que las personas que intentan venderte algo.

Esto es lo que cambia esa capa.

Sin ella, preguntas si algo es “innovador”.

Con ella, preguntas con qué datos fue entrenado, dónde falla, cuánto cuesta ejecutarlo a escala y qué decisiones automatiza realmente.

La primera pregunta puede responderse con un folleto.

Las otras cuatro no pueden responderse escondiéndose detrás de palabras bonitas.

Ese es todo el cambio.

Y requiere aproximadamente una hora por curso.

Diez cursos de MIT que vale la pena revisar.

No para volverte técnico.

Para dejar de ser rodeado en las decisiones.

1. AI 101. El punto de partida para protegerte de los vendedores.
https://t.co/C1ZTcYtNKS

2. Artificial Intelligence. Separar la realidad del humo.
https://t.co/Rmo3t3W16t

3. Foundation Models and Generative AI. Detectar promesas exageradas.
https://t.co/OVBCfSIF64

4. Introduction to Machine Learning. Financiar las apuestas correctas.
https://t.co/iPHRnrzZ30

5. Understanding the World Through Data. Hacer mejores preguntas.
https://t.co/zk3pbZqLmQ

6. Introduction to Deep Learning. Entender qué cuesta realmente millones.
https://t.co/S3ts87ukbZ

7. ML with Python. Ver si tu equipo está construyendo o dando vueltas.
https://t.co/8znJ9Luc7i

8. How to AI Almost Anything. Encontrar espacios competitivos.
https://t.co/rlSxHg04rL

9. Introduction to Algorithms. Defender decisiones de IA bajo escrutinio.
https://t.co/NPFAG5Wfs8

10. AI in K-12 Education. Ver la transformación antes de que te alcance.
https://t.co/TKtVpQQkdf

Esta es la parte que la mayoría pasa por alto.

El riesgo de seguir siendo no técnico no es solo que los proveedores te vendan humo.

Es que tu propia organización deja de traerte las decisiones importantes.

Cuando un líder no puede evaluar la IA, los equipos aprenden a rodearlo.
Predeciden y luego presentan. La firma se convierte en una formalidad.

Así es como un ejecutivo se convierte en un sello de aprobación sin darse cuenta.

La alfabetización es la forma de seguir siendo quien toma la decisión, no quien es llevado hacia un “sí”.

Así que elige un curso esta semana. Dedícale una hora.

No estás aprendiendo a construir. Estás aprendiendo a seguir estando en la sala.

[Alex Prompter @alex_prompter]

The 3 prompts I run after every voice dump:

1. The brief:
"Here's a raw voice transcript. I was thinking out loud so it's messy and out of order. Turn it into a tight brief: core idea in 2 sentences, key points grouped by theme, decisions I already made, open questions, next actions. Keep my phrasing where it's strong, cut everything else. Don't add ideas I didn't say."

2. The skeptic:
"Now read the same transcript like a skeptical investor. Where do I contradict myself? What am I assuming with zero evidence? Give me the 3 weakest points and one question for each that I'd have to answer before showing this to anyone."

3. The sorter:
"Split the transcript into: decisions made, action items with any deadlines I mentioned, ideas to park, and anything that could work as a post or hook. Flag whatever I said more than once, the thing I keep repeating is usually the thing that actually matters."

Run them in that order. 1 gives you the doc, 2 kills the weak version early, 3 makes sure nothing dies in the transcript.

[Claude @claudeai]

Claude Cowork is coming to mobile and web.

Hand Claude a task at your desk and pick up the finished work from your phone. Close the laptop and Claude keeps going.

Beta is rolling out over the next several weeks starting with the Max plan, with more plans to follow.

[peter @iampeterai]

8 AI Concepts That You Should Focus in Rest of 2026

[Jokker @0xJokker]

STANFORD ACABA DE FILTRAR GRATIS LA CLASE QUE EXPLICA COMO FUNCIONA CLAUDE Y CHATGPT POR DENTRO

La mayoria desperdicia el 90% de su potencial

Stanford te lo enseña en 1h44 minutos

Guardalo en favoritos para no perderlo 🔖

[Alex Rivas | IA @alexconia]

🚨 ÚLTIMA HORA: Claude ya puede crear una página web ULTRAPREMIUM para ti en solo 30 minutos.

Y totalmente GRATIS.

Aquí tienes 8 prompts para CREAR LA WEB DE TUS SUEÑOS:

[kaize @0x_kaize]

Claude Code just dropped "Getting Started with Loops"

This is their first official document about Loop Engineering.

Spoiler: prompt engineering didn't survive.

Here's the full guide in one post:

1. Turn-based

Every prompt you send already runs as a loop: Claude gathers context, takes action, checks its own work, and repeats until it decides the task is done or realizes it needs your input.

2. Goal-based

You define what "done" looks like with /goal, and Claude keeps iterating toward it. Every time it tries to stop, a separate evaluator model checks your condition - if it's not met, Claude gets sent back to work until the goal is reached or the turn limit hits.

3. Time-based

This is the /loop command: нou set an interval and a prompt fires on schedule. For example, every 5 minutes Claude checks your PR, addresses review comments, and fixes failing CI.

4. Proactive

Here you set up an event once, and the loop triggers itself whenever that event happens - with zero human input.

Anthropic also shared optimization advice - and this part matters most:

- Loop quality depends on system quality.
- Tokens are the real cost

None of this is magic. It's the same familiar parts - /goal, /loop, Skills, Hooks - composed into systems that run without you.

I built exactly that kind of system in my article:

Loop that wakes up at 6 AM, finds work on its own, reviews itself, and leaves ready PRs for you. Step by step, from SKILL(.)md to cron trigger:

[Lionel Mora @lionel_mora]

Following the amazing reaction to the Marble Curriculum yesterday, we've decided to make it open source 🛰️👇

Everything a child learns in primary school. 1,590 concepts. 3,221 connections across 8 subjects, from Math and Science to Computing and Life Skills. Anchored in the US and UK curriculums, standard by standard (NGSS, Common Core, DfE).

What you will find in the repo: every concept as structured JSON with its age band and the evidence a child must show to master it. Every prerequisite link marked hard or soft, with a written rationale. It's a true DAG you can compute learning paths on. Open license, you can build whatever you want with it.

Now is a unique time in history to be building in education. Getting AI and kids education right is likely one of the hardest and most important problems to crack over the next decade and we need as many smart and creative minds behind it.

We think a common solid basis, accessible to all and that can be built upon, is critical to move fast. That's why we're making this curriculum open source.

It's not perfect but we know it's a robust basis, and we believe that sharing it openly is the fastest way to progress in this field. If you're building in education, share this around you and tell us in comments if you find this useful and if you want to contribute.

We'll keep working and investing on it @withmarbleapp. Credit goes to @guillaume_boni for building this. I just made it look pretty.

Links below 👇

[Ismail Khan @learnbyIsmail]

🚨Anthropic just showed a 24-minute workshop on how to actually do prompts for Claude.

Taught by the people who built it.

Free. No registration. No paywall.

I've seen $300 courses that don't cover what they teach in the first 8 minutes.

Watch it and bookmark it now!