Focus

Artificial Intelligence

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

How do you train people on a subject that changes every week?

For the past year, we’ve been training development teams to use AI agents across the entire software development lifecycle. In that time, we’ve delivered 47 trainings to more than 500 participants. Here’s what goals companies bring to us, what sets our training apart, which moments have stuck with our trainers, and why the real work starts after the training ends.

Blog Post

The Data Dilemma, Part 2

The first part showed why agents fail on undocumented data semantics and why Data Mesh is not a realistic answer for many mid-sized companies. This second part gets concrete: a Minimum Viable Data Mesh in four layers, from an honest inventory through curated views with ODCS contracts to agent access via MCP. A purchasing agent that compares suppliers and evaluates price histories serves as a running example.

Blog Post

The Data Dilemma, Part 1

Data Mesh and the reality in mid-sized companies

Blog Post

Beyond Claude & GPT

Sure, we enjoy trying out the latest frontier model as much as anyone. But in day-to-day project work, what matters most is control over cost, data, and availability. That’s why we increasingly rely on open-weights models paired with open source tooling. Here’s what our setup looks like.

Blog Post

Trust but Sandbox

Coding agents can now do (almost) everything. They read and write every file you have access to, they run arbitrary commands, talk over the network, read your environment variables, and drive your browser with all your cookies. In the spirit of an agentic workflow, where we let agents work as autonomously as possible, that is actually the point, because it is exactly what makes them useful. Even so, we should find a way to gain more control over what happens in there.

Blog Post

RAG and Fine-Tuning

Stronger together

Blog Post

Hail Mary: Why domain knowledge cannot be extracted from experts

AI agents promise to replace weeks of requirements workshops with hours of structured interviews. The pitch is compelling. The cognitive science is not.

Blog Post

The Right Kind of Hard

More structure, more output, more exhaustion. Spec-driven development has helped me get better results from AI agents – and drained my energy in the process. About dead weight, false clarity, and hidden costs I didn’t plan for.

Blog Post

AI Features for Jira Data Center – No Atlassian Cloud Required

Our Digital Independence Day recipe for going off-cloud

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.

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.

Blog Post

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

Why BMAD won’t save you

Blog Post

Agents good in the end?

Software development consists of a constant chain of trade-offs. As long as I’ve been building systems, there have always been things you don’t do despite wanting to do them. But now, with agents, we can fulfill all our dreams and finally build everything we always wanted and had the feeling we were missing. Spoiler alert: we shouldn’t do that.

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.

Blog Post

Neuland reloaded
Blog Post

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

I want my AI coding agents to work independently, but I don’t want them to have unrestricted access to the internet. In this post, I describe how I routed all network traffic from my development sandbox through a strict proxy allowlist, allowing only a small set of explicitly approved domains. This setup finally gave me enough confidence to loosen the guardrails without constantly staying in the loop.

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

SDD describes how a system is built. SDA describes how systems fit together

Blog Post

Let’s Not Normalize Insecure AI Assistants

AI assistants like OpenClaw promise convenience, autonomy, and increasingly personalized help. But beneath that promise lies an architecture that quietly combines private data, internet access, and exposure to untrusted content—the lethal trifecta of security risk. Sandboxing and physical isolation help, but they don’t address the core problem. As we add more capabilities, the potential blast radius only grows. This post is a case for slowing down, questioning defaults, and refusing to normalize insecure architectures.

Blog Post

I sandboxed my coding agents. You should too.

LLM coding agents are extremely powerful because they can run programs on our computers using our permissions. However, this same power also makes us very vulnerable. It only takes one mistake or one prompt injection to compromise the whole system.

Blog Post

From FOMO to Focus

Clear Priorities in the AI Jungle with AI Opportunity Mapping

Blog Post

Modern Legacy (thanks to AI)

The discussion around software modernization is currently dominated by solutions involving artificial intelligence (AI) in software development. AI-assisted code completion, agentic software engineering, and other tools promise to massively accelerate development processes. The obvious thought is to apply these tools directly to existing code as well: refactorings, re-documentation, or even generated code for new features in legacy systems. Technically, much of this is already possible, and the buzzword industry is once again in full swing. Yet these code-centric, AI-driven approaches can be shortsighted for organizations tangled in a jungle of legacy systems. A modernization that focuses solely on code quickly leads to modern legacy: systems that appear technically advanced and well documented thanks to AI, but whose actual purpose has long become questionable.

Blog Post

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

Blog Post

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.

Blog Post

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.

Blog Post

Context Engineering

Powering the Next Generation of Intelligent AI Agents

Blog Post

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.

Blog Post

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).

Blog Post

Transcribing podcasts with large language models
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

How can we transform a pile of diverse documents into a high-quality knowledge base for AI systems? Document ingestion is the critical first step in any Retrieval-Augmented Generation (RAG) system—and often the reason why these systems fail.

Blog Post

Retrieval-Augmented Generation

Bridging General and Domain-Specific Knowledge

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

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

LLM-assisted Abbreviation Mining for Legacy Systems

Deciphering the Hidden Language within Old Software Programs

Blog Post

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?

Blog Post

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

Curious on how to start building with Generative AI? This guide covers the basics, from deploying local models to utilizing tools like Huggingface and LangChain. Learn to create your first AI application, manage customer feedback, and grasp key concepts such as context windows and vectorization. Discover practical examples and essential resources to help you get started.

Blog Post

How To Build a Data Product with Databricks
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.

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.

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.

Blog Post

Fairness and Artificial Intelligence

Why Metrics Are Not Enough

Blog Post

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.

Blog Post

MLOps and Model Governance

MLOps and model governance are often viewed as separate processes. And yet they rely heavily on one another. In this article we therefore propose the integration of these frameworks and explain the most important principles and technical components of MLOps and ML model governance.

Blog Post

What tracks do we leave behind with technology?

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

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.

Case Study

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

Read case study
Case Study

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

Read case study
News

INNOQ launches Data and AI Consulting Services
Case Study

SACAC optimizes the quotation process with a customized software solution

Read case study
Case Study

Gaining a competitive edge in the quotation process through Machine Learning

Read case study
Training

Designing software architectures for AI and ML systems

iSAQB® Training
Dates & Details