Digital Sapiens

¡Hola! Si quieres estar al día de todas las novedades de la comunidad Digital Sapiens OS, no te pierdas la newsletter que @higinix envía cada domingo. Suscríbete gratuitamente aquí: digitalsapiensos.com/newsletter

🚨 It's been just 88 hours since the ChatGPT Agent launched. And the Internet is already losing its mind. Here are 15 WILD ways people are using it (wait till you see #5):

Fine-tune 100+ LLMs directly from a UI! LLaMA-Factory lets you train and fine-tune open-source LLMs and VLMs without writing any code. Supports 100+ models, multimodal fine-tuning, PPO, DPO, experiment tracking, and much more! 100% open-source with 50k stars!

Shocker! Claude 4 system prompt was leaked, and it's a goldmine! The Claude system prompt incorporates several identifiable agentic AI patterns as described in "A Pattern Language For Agentic AI." Here's an analysis of the key patterns used: Run-Loop Prompting: Claude operates within an execution loop until a clear stopping condition is met, such as answering a user's question or performing a tool action. This is evident in directives like "Claude responds normally and then..." which show turn-based continuation guided by internal conditions. Input Classification & Dispatch: Claude routes queries based on their semantic class—such as support, API queries, emotional support, or safety concerns—ensuring they are handled by different policies or subroutines. This pattern helps manage heterogeneous inputs efficiently. Structured Response Pattern: Claude uses a rigid structure in output formatting—e.g., avoiding lists in casual conversation, using markdown only when specified—which supports clarity, reuse, and system predictability. Declarative Intent: Claude often starts segments with clear intent, such as noting what it can and cannot do, or pre-declaring response constraints. This mitigates ambiguity and guides downstream interpretation. Boundary Signaling: The system prompt distinctly marks different operational contexts—e.g., distinguishing between system limitations, tool usage, and safety constraints. This maintains separation between internal logic and user-facing messaging. Hallucination Mitigation: Many safety and refusal clauses reflect an awareness of LLM failure modes and adopt pattern-based countermeasures—like structured refusals, source-based fallback (e.g., directing users to Anthropic’s site), and explicit response shaping. Protocol-Based Tool Composition: The use of tools like web_search or web_fetch with strict constraints follows this pattern. Claude is trained to use standardized, declarative tool protocols which align with patterns around schema consistency and safe execution. Positional Reinforcement: Critical behaviors (e.g., "Claude must not..." or "Claude should...") are often repeated at both the start and end of instructions, aligning with patterns designed to mitigate behavioral drift in long prompts.

We’re very proud to finally announce *Appwrite Sites* 🍾 The open-source Vercel alternative. You can now use Appwrite to build everything you need on one platform, including: ✅ Auth ✅ Databases ✅ Functions ✅ Storage ✅ Messaging ✅ Realtime ✅ Hosting

Claude for prompting + ChatGPT for static generation is insane. I made all of these entirely with AI. Vibe marketing is HERE.

GPT-4o image generation for branding. this saves days of work 🤯 7 examples + prompts:

I made this concept brand in ChatGPT 4o (first time using the image tool) + @Shopify in ~2 hours. Name, packaging, tagline, testimonials, homepage copy, and product shots, all AI. Pretty wild.

Alright—it's hard to ignore the new tool in the designer's toolkit. I added a new page on my website with my perspective on using AI-generated images in branding, along with some examples of client and personal work where I have done that. link below 👇

Ok, @OpenAI GPT-4o can generate 2D character animation sprite sheets for your vibe coded video game Prompt below 👇

Hey @OpenAI o3, analyze your own system prompt and create an infographic of your capabilities TIL about "rich ui widgets" 👇

CFO Sarah Friar revealed that OpenAI is working on: "Agentic Software Engineer — (A-SWE)" unlike current tools like Copilot, which only boost developers. A-SWE can build apps, handle pull requests, conduct QA, fix bugs, and write documentation.

OpenAI CEO, Sam Altman, predict: Another major shift in software development is coming soon, as "Agentic Software Engineering" begins to take off.

¡Grandes editoriales como Facebook, New York Times y más se niegan a permitir que Apple Intelligence entrene con sus datos! ¿Qué consecuencias tendrá esto en el futuro? 😮 #Tecnología appleinsider.com/articles/24/08…

¿Dónde estará la acción de Nvidia en un año? @themotleyfool analiza y proyecta el futuro de $NVDA $GS $META en fool.com/investing/2024… #bolsa #inversiones

Increíble logro de Google: recrea versión jugable de Doom solo con IA. ¡Impresionante avance tecnológico! #IA #Doomhttps://www.eurogamer.net/google-reaches-grim-ai-milestone-replicates-playable-version-of-doom-entirely-by-algorithm

¡Accenture y Google Cloud unen fuerzas para avanzar en la adopción de IA y ciberseguridad! #innovación #tecnología seekingalpha.com/news/4145074-a…

Inversores no quedan impresionados con el superávit de Nvidia en el segundo trimestre. ¿Cuáles son las razones? Descúbrelo en este enlace: finance.yahoo.com/video/nvidia-3…

¡Polémica con el entrenamiento de IA de Apple! Grandes editores deciden no participar. ¿Qué repercusiones tendrá esto en el futuro? #IA #Apple phonearena.com/news/apple-ai-…

Los analistas revisan el precio objetivo de las acciones de Nvidia después de sus ganancias del segundo trimestre. ¡Descubre más aquí! #Nvidia #Analistas #GananciasTrimestrales thestreet.com/investing/stoc…

¡Atención a los niveles de precio de Salesforce tras saltar las acciones después de unos excelentes resultados! #Salesforce #Earnings investopedia.com/watch-these-sa…