Focus

Artificial Intelligence

Articles, podcasts, talks, and more about Artificial Intelligence.
Blog Post

Where AI Helps (and Hurts) Across Different Coding Scenarios

Where does LLM-assisted software development improve developer productivity—and where does it fall short? Instead of treating AI in software development as a one-dimensional productivity booster, we explore the question across several dimensions suggested by a Stanford-adjacent study: project maturity, task complexity, and the popularity of the programming language. The goal is to create a more realistic expectation baseline for both software developers and leaders—well away from today’s hype.

Podcast

KI-Agenten einsperren

Sandboxen und Prompt Injection

Anja Kammer
Joy Heron
Blog Post

Spec-Driven Development is Domain-Driven Design’s Impatient Cousin

Why BMAD won’t save you

Blog Post

REST Beats MCP

Instead of using dedicated APIs, agents can operate existing web applications directly. Like humans, they rely on the most consistent implementation of hypermedia and benefit from existing context, validation, and access controls. Using an expense report as an example, I show how agents can automate complex, context-dependent tasks—without having to implement new APIs.

Podcast

AI News

Wachstum bis zum Blackout. Das große StĂŒhlerĂŒcken. Die Agents.md auf dem PrĂŒfstand.

Fabian Walther
Ole Wendland
Blog Post

Agents good in the end?

A story of unleashed productivity and mental load

Blog Post

Understanding AI Coding Patterns Through Cognitive Load Theory

AI coding assistants either severely harm learning or enhance it beyond manual coding—depending entirely on interaction patterns. A recent study reveals six distinct patterns with dramatically different outcomes. We’ll examine why through cognitive psychology and explore the study’s limitations.

Article

Neuland reloaded

When Angela Merkel said in 2013 that the internet is ‘Neuland’ (uncharted territory) for all of us, the amusement online was considerable. Viewed from some temporal and substantive distance, one can now recognize a perspective from which she was indeed correct.

Blog Post

I sandboxed my coding agents. Now I control their network.

Restricting network access for AI coding agents with a proxy allowlist

Blog Post

From Vibe Coder to Code Owner

AI agents generate thousands of lines of code in no time. If you want to fully leverage their potential, you can no longer review every single line – but you’re still responsible for the software. How do you take ownership of code you haven’t fully read? The answer lies in the agent harness: a system of deterministic checks, AI reviews, and targeted human review that enforces quality instead of hoping for it.

Blog Post

Spec-Driven Architecture: When Agents Build, Architecture Must Speak

Spec-Driven Development gives agents a clear foundation for implementation. What it doesn’t solve is how a portfolio of systems stays coherent. Spec-Driven Architecture applies the same principle at the architecture level, using contracts as versioned boundaries and guarantees—enforceable in agentic workflows and in the CI/CD pipeline.

Blog Post

Let’s Not Normalize Insecure AI Assistants

How the Lethal Trifecta Makes Today’s AI Assistants Unsafe by Design

Blog Post

Die persönliche KI ist schon da

Drei Wochen mit OpenClaw auf einem Raspberry Pi und ein Blick darauf, was auf die Softwarearchitektur zukommt.

Blog Post

Fetch-Tools vs. Browser-Rendering in Agenten-Setups

„Eine URL abrufen“ klingt trivial, kann aber innerhalb eines Agenten schnell chaotisch werden. Ein Teil der Verwirrung entsteht durch Dokumentation, die nicht sauber trennt, was die einzelnen Web-Tools eigentlich leisten.

Blog Post

I sandboxed my coding agents. You should too.

My Lima VM + JetBrains Gateway setup for safe agentic development

Blog Post

From FOMO to Focus

AI is on every agenda – but where do you start? Many companies launch multiple AI initiatives in parallel, driven by the fear of missing out. The result: scattered resources, lack of prioritization, and unclear business impact. AI Opportunity Mapping provides a solution: it systematically guides you from vague AI visions to concrete, prioritized use cases with real business value. In five steps, teams develop AI opportunities in a structured way and make well-founded decisions. The result: focus instead of FOMO, clarity instead of chaos.

Article

Prompts als Programm in GPT-5

All Caps, Gebettel und Geschrei – Schnee von gestern. GPT-5 befolgt brutal jede Instruktion, selbst in versteckten NebensĂ€tzen. Und verhungert, wenn wir nicht genau sagen, was wir wollen. Welche Konsequenzen hat das fĂŒr Prompt Engineering und Agentenentwicklung?

Blog Post

Modern Legacy (thanks to AI)

Why Focusing on Code Alone Falls Short in the Age of AI

Blog Post

Modern Legacy (dank KI)

Die Diskussion ĂŒber Softwaremodernisierung wird derzeit stark von Lösungen rund um kĂŒnstliche Intelligenz (KI) in der Softwareentwicklung geprĂ€gt. KI gestĂŒtzte Code-VervollstĂ€ndigung, Agentic Software Engineering und weitere Werkzeuge versprechen, Entwicklungsprozesse massiv zu beschleunigen. Der Gedanke liegt nahe, diese Werkzeuge direkt am bestehenden Code einzusetzen: Refactorings, Redokumentation oder sogar generierter Code fĂŒr neue Funktionen in vorhandenen Legacy Systemen. Technisch ist hier bereits viel möglich und auch die Buzzword-Industrie ist hier wieder massiv unterwegs. Doch diese Code-nahen, KI-getriebenen AnsĂ€tze können fĂŒr Unternehmen, die sich mit einem Dschungel aus Legacy Systemen auseinandersetzen, zu kurzsichtig sein. Eine Modernisierung, die von Anfang an nur auf Code zielt, fĂŒhrt schnell zu Modern Legacy: Systeme, die dank KI-UnterstĂŒtzung technisch hochmodern und vorbildlich dokumentiert wirken, deren eigentlicher Zweck aber lĂ€ngst fragwĂŒrdig ist.

Article

Think Locally: On-Premise LLMs as Drivers of Competitive Advantage

Consider this scenario: Your organisation has integrated AI tools into critical business processes, your legal team has carefully reviewed data processing agreements, and your IT department has configured systems to comply with GDPR requirements. Then, a foreign court issues an order that overrides all these protections, requiring your AI provider to indefinitely retain data that should be deleted—including potentially sensitive corporate information shared by your employees.

Blog Post

Software Analytics going crAIzy!

From analyzing problems in software systems to working solutions with AI

Article

The Sovereignty Trap: Between Tiananmen and Trump

We talk a big game about values and privacy, yet depend on AI that either denies historical atrocities or could cut us off tomorrow. Is this what Europe gets for dropping the ball on high tech infrastructure?

Blog Post

Swiss Army Knife for Salesforce: LLM with In-Memory Database

Large Language Models (LLMs) struggle with transforming large datasets, particularly when performing aggregations that require calculations - essentially anything where you’d normally use GROUP BY or ORDER BY in SQL. But what if we need to retrieve extensive data from a system and process it with an LLM, but only have an API with limited functionality available? In this blog post, I present a solution I’ve implemented and tested.

Article

From Data Graveyards to Knowledge Landscapes
Blog Post

AI — Behind the Buzzword Garbage

Tired of AI hype? Me too. But beneath the buzzwords lies real value for developers. Tools like Claude Code save me hours on routine tasks, freeing me to focus on what matters: understanding problems and building the right solutions. It’s not magic—it’s practical support that makes development faster without replacing our core skills.

Blog Post

First Agile, Then Agentic

Agentic AI is supposed to accelerate software development. But new technologies can only reach their full potential when organizations adapt their structure, processes, and culture. Most organizations today are not yet able to truly benefit from faster software development. The prerequisite for this are the capabilities shaped by the agile and DevOps movements.

Article

Asset Administration Shell und Model Context Protocol

Freund oder Feind?

Blog Post

Context Engineering

Have you ever wondered how chatbots, copilots, or virtual assistants seem to know just the right things to say — or sometimes, spectacularly fail? The difference often comes down to one crucial skill: context engineering.

Blog Post

Plattformen fĂŒr AI: Daten und APIs als digitales Fundament

Eine aktiv gemanagte Plattform von Daten und Diensten ist die Grundlage fĂŒr agentenfĂ€hige Systeme. Nur wer Daten und Dienste gemeinsam zugĂ€nglich macht, schafft die nötige Basis fĂŒr AI. Agenten brauchen klar beschriebene Datendienste – und eine Plattform, die diese zur VerfĂŒgung stellt.

Blog Post

Das Versprechen der Agenten
Blog Post

Macht die KI die Softwareentwicklung effizienter?

Millionen Zeilen Code fĂŒr ein einfaches “Hallo Welt”? Dieser Post beschreibt, wie aufgeblĂ€hte Open-Source-Projekte unsere Softwareentwicklung verlangsamen und wie KI diesen Trend umkehren könnte. Ist unsere Software wirklich so effizient, wie wir wollen? Eine Analyse, die zum Nachdenken anregt.

Blog Post

Agenten - Kreise - Firmen

In Teil 2 habe ich Ihnen versprochen, die probabilistische Weltsicht zu beleuchten, mit der Machine Learning Experten schon immer vertraut sind, und die uns allen mit agentischen KI-Systemen bevorsteht. Wir befassen uns damit, was genau in unserem Fall das Lernende System ist, welche neue Rolle dabei Menschen einnehmen, und wie die Vision des CEO von Microsoft Satya Nadella zur Firma der Zukunft aussieht.

Blog Post

Wertschöpfung in Zeiten agentischer KI-Systeme
Blog Post

Zu blöd fĂŒr Vibe Coding?

Vibe Coding, also die Generierung von Code mit Hilfe von AI, gewinnt zunehmend an PopularitĂ€t. Mit Claude Code hat Anthropic ein Tool entwickelt, welches mir zum ersten Mal das GefĂŒhl gegeben hat, dass diese Art von Coding damit auch fĂŒr mich funktionieren könnte. Ich hab’s ausprobiert, mit einer Problemstellung, die dafĂŒr wie gemacht zu sein schien. Lest nach, wie es mir dabei ergangen ist und welche Erkenntnisse ich gewonnen habe. Spoiler: es hat funktioniert, am Ende, irgendwie, aber war das noch Vibe Coding?

Blog Post

DatensouverÀnitÀt unterwegs: OpenWebUI trifft Ollama im VPN

In diesem Beitrag zeige ich, wie ich LLMs mit Ollama auf einem mobilen MacBook betreibe – und dank VPN und OpenWebUI von ĂŒberall darauf zugreifen kann. Sicher, und selbst gehostet. Schritt fĂŒr Schritt zum persönlichen KI-Setup – performant, privat und unterwegs verfĂŒgbar.

Podcast

MCP – Model Context Protocol

Der Universalstecker fĂŒr KI-Modelle

Robert Glaser
Ole Wendland
Article

Beyond the hype: An engineer’s journey into ReBAC and AI with the Model Context Protocol

In this article, I share my experiences on my journey into the AI world. During this journey, we’ll build our own Model Context Protocol (MCP) Server using C Sharp, learn about access management with relationship based access control (ReBAC) on the way, and in the end I’ll provide my thoughts on the current state of AI and MCP, focusing on security and UX.

Article

Building Standardized AI Tools with the Model Context Protocol (MCP)

In this article, we’ll explore the integration of Large Language Models and systems built on top of them. The key concept in this space recently is the Model Context Protocol (MCP).

Podcast

KI Agenten

Von Workflows zu autonomen Systemen

Robert Glaser
Ole Wendland
Blog Post

Transcribing podcasts with large language models

Automatically transcribing podcasts with multimodal AI sounds like a no-brainer, but the devil is in the details.

Blog Post

RAG: The Architecture of Reliable AI

How can we ensure that AI systems are precise, transparent, and always up to date? All Large Language Models (LLMs) have a cut-off date where their world knowledge ends. And they know nothing about your company’s internal information. Even the leading models still have hallucination rates we can’t completely ignore. Yet they offer enormous potential for productivity, efficiency, and creation. Retrieval-Augmented Generation (RAG) addresses exactly this issue: LLMs are enhanced through targeted information retrieval.

Blog Post

Document Ingestion

The Foundation of a RAG System

Blog Post

Retrieval-Augmented Generation

How do we handle situations where an LLM-driven system requires highly specialized enterprise knowledge that wasn’t included in the original model training? While LLMs have enormous potential to answer generic queries based on their comprehensive knowledge from training, they show limitations when it comes to current, specialized, or verified information.

Podcast

DeepSeek R1

Ein Wal liegt am Strand

Robert Glaser
Ole Wendland
Article

LLMs mit Spring AI integrieren

Spring AI stellt sich vor

Blog Post

Better RAG With Hybrid Search

One component of RAG (retrieval-augmented generation) is the retrieval. In other words: we have to solve a search problem. Reading articles about RAG, one can get the impression that vector search is the essential or even the only piece of the puzzle. In this blogpost you’ll see why this does not bring us close enough to the goal.

Blog Post

When the Worldview is Shifting

Large Language Models (LLMs) can feel like magic. We observe them and imagine that they work in a certain way. Then something surprises us, shattering the illusion and fundamentally reshaping our understanding. In this blogpost I’ll show you one such surprise from my own experience, and I believe that your worldview will have changed at the end.

Blog Post

LLM-assisted Abbreviation Mining for Legacy Systems

Deciphering the Hidden Language within Old Software Programs

Podcast

INNOQ Technology Day

Programm und Behind the Scenes

Sven Johann
André Deuerling
Isabel König-Wingen
Article

Generative AI: The End of “Too Expensive” in Business Software?

”We’ve found no use cases.” — No, you’ve got too many. Let’s take a look at how Generative AI will change business software beyond chatbots. What features might become possible that weren’t before, or were just too expensive?

Article

Here’s All You Need To Know To Start Building With Generative AI
Article

How To Build a Data Product with Databricks

In today’s data engineering, the focus is primarily on developing modular data products. This article outlines the advantages of modularity over monolithic data pipelines and explains, step-by-step, how to develop data products using Databricks – from defining a data contract to creating and implementing Databricks Asset Bundles, setting up a CI/CD pipeline, and publishing metadata.

Podcast

KI-unterstĂŒtzte Entwicklung

ChatGPT im tÀglichen Einsatz

Anja Kammer
Isabel König-Wingen
Podcast

RAG

Abfragen und Bergen von Wissen

Robert Glaser
Ole Wendland
Podcast

Enterprise Search mit Vektordatenbanken

Was Vektordatenbanken anders machen als der Suchindex

Sven Johann
Marco Steinke
Podcast

AI Prompting

Kontext ist Gold

Robert Glaser
Anja Kammer
Blog Post

Entwickeln mit ChatGPT

Wie KI meine Programmierarbeit revolutioniert

Blog Post

Lokale LLMs mit Ollama und Spring AI nutzen

Egal, ob wir wollen oder nicht, um AI und speziell Large Language Models (LLM) kommen wir aktuell nicht herum. Mich schrecken solche Hypes zwar aus Reflex eher ab. Allerdings sieht es so aus, als wĂŒrde von diesem Hype mehr bleiben als vom letzten, der Blockchain. Deshalb wollen wir uns in diesem Post einmal anschauen, wie man ein LLM lokal aufsetzen kann und dieses mittels Spring AI in eine Spring Boot-Anwendung einbinden kann.

Blog Post

A natural language calculator

In my prior post I’ve written about how to run a chat with a large-language-model on your PC. This time I want to focus on scripting this with Node.js and letting the AI- and the “normal”-world interact with each other.

Podcast

Large Language Models

VerÀndern sie alles?

Stefan Tilkov
Marcel Weiß
Blog Post

AI Tools in Business Environments

Currently, the importance of AI tools is growing at a breathtaking pace and has also gained importance in the general public. More and more companies and organizations are relying on the advantages of artificial intelligence to improve their processes, increase their productivity, or better serve their customers. AI tools are able to reliably analyze data, recognize patterns, and make predictions that can already surpass human abilities in some areas. This makes them a valuable tool for optimizing business processes and developing innovative products and services. It’s no wonder that the demand for AI tools has exponentially increased in recent years and will continue to grow.

Podcast

Women in Tech: Larysa

Eine Frage des Outfits

Dr. Larysa Visengeriyeva
Stefanie Heinrich
Blog Post

Running an AI Chatbot on Your Own PC
Blog Post

How to use Apple Shortcuts to integrate GPT-4o in macOS and iOS

Apple Shortcuts is a powerful app that lets you create custom workflows with multiple steps using your apps and content. You can also use it to interact with web services and APIs, such as OpenAI’s Chat Completions API for GPT-4o, which can generate text completions for any prompt or task. Yup, the thing that’s behind ChatGPT.

Blog Post

How AI will replace my job

In late 2022, I decided to try to use ChatGPT, an AI language processor, to do some of my daily software development work. Now, only a few weeks later, I am convinced AI might soon do most of my current work, at least measured by hours.

Article

KI-Systeme: MLOps, Model Governance und Explainable AI sichern robusten Einsatz
Article

Fairness and Artificial Intelligence

Classical software testing cannot simply be transferred to AI. Model governance and internal audits are required to ensure fairness.

Article

Ethics and Artificial Intelligence

Artificial intelligence is forcing its way into many fields of application. Now it is important that it works in a responsible, secure, and transparent way. The regulation of AI systems is a legal, societal, and technical topic that demands broad awareness and that will become increasingly important in the years to come.

Article

MLOps and Model Governance
Blog Post

Das Test-driven Development fĂŒr eine Conversational AI

AnlĂ€sslich meines kĂŒrzlichen Wechsels vom Student zum Consultant schreibe ich in diesem zweiten Blogpost ĂŒber die Thematik meiner Masterarbeit.

Article

Machine Learning Security – Teil 2

ML kommt immer mehr in sensiblen Entscheidungssystemen zum Einsatz - z.B. in autonomen Fahrzeugen, in der Gesundheitsdiagnostik oder der KreditwĂŒrdigkeitsprĂŒfung. Dies bringt nicht nur neue Möglichkeiten, sondern auch neue Schwachstellen mit sich, die gezielt von Angriffen ausgenutzt werden können. In Teil 2 dieser Artikelserie beschĂ€ftigen wir uns mit verschiedenen Angriffstypen in der ML-Security-Landschaft und den dazugehörigen LösungsvorschlĂ€gen.

Article

Machine Learning Security – Teil 1

Eine neue Herausforderung

Security Podcast

Machine Learning Security

„Aus großer Kraft folgt große Verantwortung”

Simon Kölsch
Isabel BĂ€r
Podcast

Technologiemonster

Welche Konsequenzen hat unser Handeln?

Robert Glaser
Lucas Dohmen
Article

What tracks do we leave behind with technology?

How monsters can teach us about responsibility: INNOQ Digital Art Edition 02

Article

MLOps: You train it, you run it!

Data Science, Machine Learning (ML) und Artificial Intelligence haben in den letzten Jahren einen wahren Hype ausgelöst und viel Aufmerksamkeit in der Industrie bekommen. Man versucht mit Machine Learning Methoden entweder die ProduktivitÀt der Nutzer oder die InteraktivitÀt der Applikation zu steigern. Zahlreiche Data Science Teams verbringen ihre Zeit damit Machine Learning Modelle zu trainieren. Allerdings beobachten wir zwei Arten von Problemen, die in der Praxis entstehen. Entweder schafft es die Mehrheit der ML Modelle nicht in ein Softwareprodukt eingebunden zu werden oder das Model Deployment nimmt zu viel Zeit in Anspruch.

Podcast

MLOps

Entwurf, Entwicklung, Betrieb

Robert Glaser
Dr. Larysa Visengeriyeva
Article

Machine Learning Daten in den Griff bekommen

Mehrdimensionale Arrays fĂŒr Machine Learning

Article

Pragmatisch zum Praxiseinsatz von Machine Learning in der Cloud

Die Anzahl von Publikationen zu Computer Vision, Natural Language Processing (NLP) oder Reinforcement Learning ist heutzutage gewaltig. Dabei widmen sich die meisten ausschließlich dem Training. Doch oft mĂŒssen Data Scientists auch beim Betrieb ihrer Modelle mitwirken. DafĂŒr braucht es einen pragmatischen und unaufwĂ€ndigen Weg.

Blog Post

Handling German Text with torchtext

There is a growing list of tools that are ready to be used with non-English texts. We show common ways to integrate them in torchtext and use their language-specific options.

Article

Vorgehensweise fĂŒr maschinelles Lernen als Orientierung

Werkzeugneutrale EinfĂŒhrung

Podcast

Deep Learning

TrÀumen Maschinen von elektrischen Daten?

Lucas Dohmen
Michael Perlin
Talk
Talk

Agentic Software Modernization: Back to the Roots

iSAQBÂź Software Architecture Forum 2026 / 14:45 - 15:45

Talk
Talk

Code wird billig, Architektur entscheidend.

iSAQBÂź Software Architecture Forum 2026 / 09:15 - 10:00

Talk
Talk

Slop oder Top: Von Vibe Coding zu AI Aided Software Engineering

bettercode() GenAI Summit / 11:15 - 12:00

Talk
Talk

Datenarchitektur: Das neue RĂŒckgrat moderner Software?

JAX 2026 / 11:30 - 12:30

Talk
Talk

Agentic Coding in the Wild: Wie schlÀgt sich Agentic Coding in der Praxis?

Java User Group CH – Luzern / 18:15 - 19:15

Talk
Talk

Agentic Coding in the Wild: Wie schlÀgt sich Agentic Coding in der Praxis?

Java User Group CH – Basel / 18:15 - 19:15

Talk
Talk

No Constrain, No Gain: Structure for the Agent, Clarity for the Human, Flexibility for the Business

Agentic Software Engineering Night Berlin / 20:00 - 20:45

Talk
Talk

From Vibe Coder to Code Owner

Agentic Software Engineering Night Berlin / 18:30 - 21:30

Talk
Talk

KI- und MCP-Security

Agentic Software Engineering Night Hamburg / 19:00 - 19:45

Talk
Talk

Der Job verÀndert sich. Ich auch. Und das ist okay.

Agentic Software Engineering Night Hamburg / 20:00 - 20:45

Talk
Talk

Agentic Software & Security: Wer kontrolliert eigentlich wen?

INNOQ Technology Lunch (April) / 12:15 - 13:00

News

IT-Security bei INNOQ: Sichere Software beginnt mit durchdachter Architektur
News

Neu: Agentic Engineering Accelerator Programm
News

Jetzt anmelden: INNOQ Technology Day 2025
News

Neues Training: Agentic Software Engineering
News

INNOQ Fragezeit startet am 31. Juli
News

INNOQ Technology Day am 20. November 2025
Case Study

Data Governance without handbrakes: How AI accelerates time-to-value in Data Mesh

Read case study
News

Neues iSAQBÂź-Modul: Softwarearchitektur fĂŒr KI-Systeme
News

Neuer Primer: Retrieval-Augmented Generation (RAG)
Case Study

Answers instead of search results:
Sprengnetter unlocks real estate expertise with Generative AI

Read case study
News

Jetzt anmelden: INNOQ Technology Day 2024
News

INNOQ Technology Day am 5. Dezember 2024
News

Now Live: The Women+ in Data and AI Festival Schedule
News

Neues Training: GenAI fĂŒr Entwickler:innen
News

INNOQ launches Data and AI Consulting Services
News

Neu bei INNOQ: Beratung und Entwicklung im Bereich Data und AI
News

INNOQ Technology Day 2023 am 13. November
News

Women+ in Data and AI Summer Festival 2024
News

Technology Day 2023: am 13. November ist es wieder soweit!

News

INNOQ Technology Briefing

News

Women+ in Data and AI Summer Festival
News

Neuer Primer: MLOps
News

Neues Training: Domain-driven Design fĂŒr Machine-Learning-Produkte
Case Study

Gaining a competitive edge in the quotation process through Machine Learning

Read case study
Case Study

SACAC optimizes the quotation process with a customized software solution

Read case study
Training

Designing software architectures for AI and ML systems

iSAQBÂź Training
Dates & Details