Guten Morgen!
Erstmal danke ich euch für all die lieben Worte – sie bedeuten mir unglaublich viel.
Es ist das erste Mal, dass ich diesen Screenshot jemandem außerhalb von Familie und Freundeskreis gezeigt habe.
Hier habe ich noch eine kleine Testfrage aus dem Modul „Beziehung“ für euch.
Sie zeigt ganz gut, warum mein System anders funktioniert als viele andere.
#Selbstfürsorge #ADHS #PTBS #Alltagskämpfe #Appentwicklung #Neurodivergenz
#appentwicklung #ki #ai #PromptEngineering
#appdevelopment

Hier sage ich lieber vorab dazu. DIES IST EIN TESTDIALOG.
Aber * Eltern sein* in neurodiversen Familie... hat eigene Herausforderungen. Auch dafür wollte ich einen Raum schaffen.
Ohne Schuld und ohne Druck.

#Selbstfürsorge #ADHS #PTBS #Alltagskämpfe #Appentwicklung #Neurodivergenz
#appentwicklung #ki #ai #PromptEngineering #appdevelopment

Ich weiß, dass ich Dinge zu tun hätte – aber schon der Gedanke daran macht mich müde.
Manchmal hilft kein Plan, keine Liste.
Nur die Erlaubnis, da zu sein, wo ich gerade bin.
Auch das wollte ich in einer App umsetzen: eine echte Nicht-Können-Toleranz.
Und ich glaube, es ist mir gelungen.

#Selbstfürsorge #ADHS #PTBS #Alltagskämpfe #Appentwicklung #Neurodivergenz
#appentwicklung #ki #ai #PromptEngineering
#appdevelopment

»The Hottest #AIJob of 2023 Is Already #Obsolete: #Promptengineering, a role aimed at crafting the perfect input to send to a #LLM, was poised to become one of the hottest jobs in #AI.« https://www.wsj.com/articles/the-hottest-ai-job-of-2023-is-already-obsolete-1961b054?eicker.news #tech #media #news

This is worth listening to if you have an interest in using AI to code, if you weren't already into coding.

#Vibecoding i believe he said...

https://open.spotify.com/episode/1gI67au7dI08hYVNTwBxs6?si=1a2f02b6d16341ae

Dieses kleine System glaubt nicht daran, dass man sich „zusammenreißen“ muss.
Es glaubt daran, dass man manchmal einen Anfang braucht, den man selbst nicht finden kann.
Und dass das okay ist.
Es wird nichts an dir repariert.
Es wird nichts von dir erwartet.
Es wird einfach anerkannt, dass manchmal schon ein Licht anschalten ein Anfang ist.
#appentwicklung #ai #neurodivergent #programming #PromptEngineering #neurodivers #adhs
#ki

All Major Gen-AI Models Vulnerable to ‘Policy Puppetry’ Prompt Injection Attack https://www.securityweek.com/all-major-gen-ai-models-vulnerable-to-policy-puppetry-prompt-injection-attack/ #ArtificialIntelligence #PromptEngineering #AIjailbreak #AI

All Major Gen-AI Models Vulnerable to ‘Policy Puppetry’ Prompt Injection Attack https://www.securityweek.com/all-major-gen-ai-models-vulnerable-to-policy-puppetry-prompt-injection-attack/ #ArtificialIntelligence #PromptEngineering #AIjailbreak #AI

I need #GenAI help who's good at #PromptEngineering and ready to help celebrate our favorite president #Trump #Musk Here's what I have so far... what'm I doin wrong‽

> donald trump and elon musk high fiving,
> while they spit roast the statue of liberty,
> musk is pulling her head onto his tiny hog,
> trump is resting his gut on the arch of her back,
> the men look winded, sweaty, but gleeful,
> lady liberty struggles as americans look away

GPT Decision Matrix Hack

Want better AI decision analysis? Try this trending prompt template:

"I need to decide on [topic]. Create a decision matrix comparing [options] using these criteria: [list 3-5 factors]. Score each 1-5 with brief justification, then calculate weighted totals prioritizing [key criteria].

I used this for house project and discovered considerations I'd missed!

What would you analyze with this?

Created a new @yousearchengine Prompt

Link: https://github.com/spsanderson/LLM_Prompts/blob/main/Horticulturalist.md

Here is an #AI engine hack you probably didn't even knew you needed.

I've just discovered it and I would like to share.

I run a "pre-prompt" on #ChatGpt.
A kinda dashboard on my sessions and recently I felt compelled to enhance it.
Previously it was in "Settings" "persistent prompt" or some such.
Sammy renamed the field to "Tell us more about yourself". But you can keep it as a prompt.
E.g. Count the words in my prompt.

That's Hack 1.

Hack 2 is the goodie.
It's only limited to 1500 characters, and if you want more, you're screwed, even if you make it super concise.

So, because #LLM does not care what language it works in, I asked it to use Asian language and it came up with #Chinese/ #Japanese (Japneese?) hybrid (or so it says).

It still formats the response in English because I instructed it. But it's super dense and way below the 1500 so I can add more instructions should I want to.

I've set it to update the session status every 10 prompts.

"This course is intended to provide you with a comprehensive step-by-step understanding of how to engineer optimal prompts within Claude.

After completing this course, you will be able to:

- Master the basic structure of a good prompt
- Recognize common failure modes and learn the '80/20' techniques to address them
- Understand Claude's strengths and weaknesses
- Build strong prompts from scratch for common use cases

Course structure and content

This course is structured to allow you many chances to practice writing and troubleshooting prompts yourself. The course is broken up into 9 chapters with accompanying exercises, as well as an appendix of even more advanced methods. It is intended for you to work through the course in chapter order.

Each lesson has an "Example Playground" area at the bottom where you are free to experiment with the examples in the lesson and see for yourself how changing prompts can change Claude's responses. There is also an answer key.

Note: This tutorial uses our smallest, fastest, and cheapest model, Claude 3 Haiku. Anthropic has two other models, Claude 3 Sonnet and Claude 3 Opus, which are more intelligent than Haiku, with Opus being the most intelligent.

This tutorial also exists on Google Sheets using Anthropic's Claude for Sheets extension. We recommend using that version as it is more user friendly."

https://github.com/anthropics/courses/tree/master/prompt_engineering_interactive_tutorial

“Vibe coding” is a new AI-driven coding style coined by Andrej Karpathy (Feb 2025), where natural language prompts to LLMs replace manual programming. It shifts focus to guiding, testing & refining AI-generated code — no deep coding skills required.
https://en.m.wikipedia.org/wiki/Vibe_coding
#VibeCoding #AI #LLM #Karpathy #AItools #PromptEngineering #SoftwareDevelopment #FutureOfCoding #AIDev

https://www.europesays.com/uk/30264/ 5 Powerful AI Prompts That Can Boost Any Business Idea #AI #AIPrompts #ArtificialIntelligence #Business #BusinessIdeas #BusinessPlanGenerator #Entrepreneurship #GenerativeAI #PromptEngineering #PromptIdeas #UK #UnitedKingdom

Ich habe einfach gemacht.
Ich bin keine Entwicklerin. Keine Tech-Frau. Keine Forscherin. Ich komme aus der Verwaltung. Und trotzdem habe ich in 19 Tagen begriffen, wie man eine KI zum Atmen bringt. Wie man sie so strukturiert, dass sie nicht nur antwortet – sondern zuhört.
Und vielleicht braucht es dafür nicht mehr als eine Bürokauffrau mit einem unverschämten Maß an Liebe zum Detail.
#adhs #ai #ki #appentwicklung #tech #programmierung #SystemEngineering #PromptEngineering

Want to teach AI to recognize your personal writing style?

Try this prompt pattern:
Here are examples of my writing style:
[3-4 samples of your writing]

Please rewrite the following in my style while preserving all information:
[content to transform]

Consistent brand voice across all communications, even when delegating content creation.

The secret: It leverages few-shot learning without fine-tuning costs!

A Field Guide to Rapidly Improving AI Products – O’Reilly

This article subverts traditional tools-centric AI development by revealing how a focus on qualitative error analysis can uncover actionable, domain-specific weaknesses.

Its analysis, addresses both strategic and operational challenges while acknowledging the evolution of evaluation criteria in AI systems.

https://www.oreilly.com/radar/a-field-guide-to-rapidly-improving-ai-products/

Want to force LLMs to return perfectly structured data?

JSON mode is the secret weapon for developers integrating AI into applications! Instead of parsing messy text, get direct structured output.

---
You MUST respond ONLY in JSON format:
{
"answer": "your response",
"confidence": 0-10,
"sources": ["list", "your", "reasoning"]
}
---

This technique saved me 10+ hours of parsing code last week alone. Your backend will thank you!

https://www.europesays.com/uk/21435/ Google unveils new prompt engineering playbook: 10 key points on mastering Gemini, other AI tools | Technology News #AI #AIChatbots #AIGeneratedOutputs #ArtificialIntelligence #GenerativeAI #GoogleGemini #GoogleVertexAI #JSONOutputFormatAI #LLMs #ModelConfigurationsAI #PromptEngineering #PromptingTechniques #Technology #TextPrompts #UK #UnitedKingdom