Blog Posts

Long-form articles and blog posts
A hand-made paper collage on a sheet of old ledger paper. On the right, a man in a suit walks away through an open doorway into daylight. His shadow stretches back across the floor to the left, but it is not dark: it is assembled from hundreds of cut-out paper receipts and forms. The shadow has not followed him. It sits hunched over a desk, still working, while a bright orange till roll unspools across the floor. A small black-and-tan dachshund in a red bandana sits in the paper pile, a receipt stuck to his head, looking at the viewer.

I was the tax office’s unpaid clerk, until I delegated it to an AI

| In AI
| 15 minute read

Every summer, a shoebox wins.

For about fifteen years, preparing my German tax return has cost me a weekend plus a few evenings, and quite a lot of swearing in between. Not the tax return itself: I have a Steuerberaterin for that, and she is excellent. What costs me the weekend is everything that has to happen before she can start. Remembering which receipts exist. Finding them. Naming them. Sorting them into the right folders. Building the spreadsheets that explain the numbers, and the notes that explain the spreadsheets.

This year I handed most of that to an AI. It still took a couple of days. But for the first time since I have been filing, it didn’t suck, and the result was better than anything I had produced on my own.

Five open books on a wooden table in a sunlit library, each sending a coloured ray of light up into a glass orb, which combines them into a single white beam.

perplexity-agent-mcp: an open-source MCP server for Perplexity’s Agent API

| In AI
| 7 minute read

I regularly use the Perplexity API, whether from the command line or by letting Claude use it through an MCP, as a second search engine for its research. That’s why I paid attention when, about a week ago, an email from Perplexity landed in my inbox with the subject line “Your Sonar workload runs better on Agent API.” The pitch: their newer Agent API beats the classic Sonar models I’d been using, on both quality and cost.

Their own benchmarks put the cheapest Agent preset at roughly twice the accuracy of Sonar for about the same money, with the deeper presets reaching a level Sonar can’t touch (79.67% vs. 29.10% on BrowseComp for the high preset against Sonar Deep Research, at about 27% higher cost per query).

Marketing numbers, of course, but the direction made sense, and Perplexity is clearly steering everyone from Sonar to the Agent API (though they did also say Sonar isn’t going away).

I wanted to try it out and switch to the better, more efficient engine. It sounded easy, only it wasn’t.

A bearded male chef wearing glasses, a white chef’s coat, and chef’s hat stands in a modern kitchen with modern kitchen appliances visible behind him. He holds a spoon in one hand and a recipe card in the other. Four small white robot assistants with round heads, yellow eyes, and gold accents wearing chef’s hats stand around him, helping with cooking tasks. The kitchen features stainless steel countertops, hanging copper pans, open flames on the stove, fresh ingredients, and recipe cards clipped to an overhead rail. The scene depicts a futuristic collaborative cooking environment blending human expertise with robotic assistance.

Why working with AI agents is so tiring (and what to do about it)

| In AI
| 9 minute read

Earlier this week, I spent five hours at the WERK1 coworking space in Munich, moving three projects forward at once: the brain dump that would become this blog post, a contribution to an open source project, and a redesign of my blog’s analytics. My AI agents did the heavy lifting. Whenever one of t…

A cozy indoor scene showing a decorative brass birdcage with one yellow and green budgie inside, along with a mirror where the bird’s reflection is visible, a water dish, and a perch. The cage sits on a wooden table next to a potted plant. To the left, a window overlooks trees with two birds perched on a branch singing. A stack of books and part of an armchair are visible in the foreground, and beige walls with vertical paneling complete the warm, homey setting.

You’re in the driver’s seat: will you choose to stay awake with AI?

| In AI
| 16 minute read

„Nah ist / Und schwer zu fassen der Gott. / Wo aber Gefahr ist, wächst / Das Rettende auch.” — Friedrich Hölderlin, Patmos (1803)

(Near is, / and difficult to grasp, the god. / But where danger is, grows / the saving power also.)

A few weeks ago, I wrote a post titled Thinking with Claude: why Cyborg writing works better than Centaur writing. In the LinkedIn comments, a reader named Christopher pushed back: Leaning on AI to help me write, he argued, was training my brain to “surrender executive function to external sources.” Examples: freed prisoners overwhelmed by everyday choices, ex-military struggling with civilian life, people leaving long-term institutional care. The source doesn’t matter, he wrote. The offloading pattern is the same.

While my initial answer was along the lines of “using your executive function is a choice”, I think it’s not that simple, after all, and I owe Christopher a real answer. In fact, his comment pushed me into a weeks-long thinking journey, during which I looked for research, came across interesting related articles, and reflected a lot on my own AI use. I also jammed with Claude, and yes: I used the very tool Christopher warned me about to think harder about his warning. We’ll get to that contradiction later.

A fantastical illustration of a serene pond ecosystem at sunrise, surrounded by vibrant wildflowers and lush vegetation. In the water, a frog holds a lantern on a lily pad while a bee hovers nearby. Golden glowing lines connect various insects—including dragonflies, a beetle, and a snail—representing an interconnected network or ecosystem. Solar panel-topped greenhouse domes are visible in the misty background among trees. The warm, ethereal lighting and whimsical art style suggest themes of nature, technology, sustainability, and ecological harmony.

Beyond coding agents: how AI rewires your organization

| In Career & Growth
| 13 minute read

TL;DR: AI made the coding part of software fast, and that speed didn’t create chaos so much as expose where your organization was already slow: the hand-offs between teams. The fix isn’t a better tool, it’s better wiring. We rewired Dev and Ops once already and called it DevOps. The next wall, between business and engineering, is coming down the same way. Here’s how to see it, and one thing you can do tomorrow.

In a Marvel comic-style recreation of the nucleological center from *The Fifth Element*, a futuristic command hub features glowing orange controls and blue monitoring stations. A man in a blue suit with a red tie grimaces as a woman leans in to whisper to him, with a speech bubble reading ‘You have 19 more meetings after this one.’ Behind them, several crew members in blue uniforms work at stations, while a large illuminated circular portal or control interface dominates the center wall with its warm orange glow. The scene captures the iconic sci-fi aesthetic of the original setting, reimagined in bold comic book art style.

Welcome Claude, my Chief of Staff

| In AI
| 28 minute read

Most people use AI like a search engine with extra steps. Ask a question, get an answer, move on. I did too, until I accidentally stumbled into something completely different.

Over the week, I’ve turned Claude into my personal Chief of Staff. What’s a Chief of Staff, you might ask? Think of these people who follow executives around, whispering important things into their ear, and making sure nothing falls through the cracks.

If you’ve watched The Fifth Element (go watch it, if you haven’t, it’s a great movie!), you know the scene: the president’s aide leans in and whispers something absurdly specific like “You have 19 more meetings after this one.” That’s essentially what Claude does for me now. Except I’m not a president. And thankfully, I have less meetings. I’m a freelance consultant living near a forest in Bavaria.

This post is a bit of a recap: how it happened, what the system actually looks like, three lessons I learned along the way, and, at the end, an unedited interview where I asked Claude seven questions about what it’s like to be my Chief of Staff, with some interesting insights. Don’t miss it!

A screenshot from a video showing a thermal POS printer displaying a list of AI-related examples. The printout shows items 6 and 7, including ‘Shell Game Podcast’ and ‘This Video Itself,’ with descriptive bullet points about each. On the right side is a close-up of Constantin, looking directly at the camera with a slight smile. Large purple text at the bottom reads ’7 ACTUAL EXAMPLES.’ The overall layout suggests this is from a YouTube video discussing real-world applications or examples of AI technology.

Video: What I actually do with AI every day

| In AI
| 2 minute read

Here’s a video I just uploaded to YouTube, walking through 7+1 examples of how AI has woven itself into my daily life — from the mundane to the surprisingly useful. No demos, no hypotheticals, rather what an actual week with AI looks like for me.

A futuristic landscape featuring a glowing neon blue digital grid flowing like a river through a vast valley surrounded by golden-lit mountains. The grid pattern illuminates with cyan electric lines against dark terrain, with scattered rocks and golden vegetation visible along the valley floor. A bright sun sets on the horizon, casting warm golden light across jagged mountain peaks, while stars dot the darkening sky above. The scene blends natural mountain terrain with digital technology aesthetics.

How to stop engineering prompts and start delegating

| In AI
| 18 minute read

TL;DR: The best “prompt engineering” technique isn’t engineering at all—it’s delegation. Transfer intent, not instructions. For quick tasks, tell the AI what success looks like and why (the Intent Prompt). For complex work, equip it with context, deliverables, and decision principles so it can navigate on its own (the Delegation Brief). You already have these skills. This post helps you apply them to AI.

Interior of a grand steampunk-gothic cathedral with ornate wooden architecture, glowing brass gears and mechanical details on the walls, illuminated by golden lights and lanterns. A massive arched doorway in the center glows with bright golden light and is framed with ornamental designs. Stained glass windows with colorful patterns are visible at the top, and a bright blue sky shines through the open doors, creating a dramatic contrast with the warm interior lighting. The floor features an intricate geometric stone pattern.

From scarce code to abundant builders

| In Career & Growth
| 7 minute read

According to media pundits, software stocks are “crashing”, and there’s a “SaaSpocalypse” going on. Recently, Noah Smith wrote about “The Fall of the Nerds,” painting a picture of software engineers as the new master weavers: skilled artisans about to be displaced by AI-powered looms. Meanwhile, I’m…

Abstract digital landscape with layered purple mountain silhouettes and flowing data streams of binary code and wireframe elements against a gradient sky, visualizing the intersection of technology and nature.

You don’t need permission to get promoted

| In Career & Growth
| 7 minute read

After 27 years in tech—working at Sun Microsystems, Oracle, and AWS, then coaching dozens of colleagues through promotions from both sides—I’ve learned something that most people get backwards: Your career belongs to you. Not your company. Not your manager. Not the promotion committee. Everything el…

A silhouetted figure stands in the center of a futuristic corridor with neon pink and blue glowing lines on the ground and vertical light beams extending upward into the distance, creating a symmetrical tunnel effect against a gradient background transitioning from blue at the top to magenta and purple on the sides.

The explore/exploit guide to your first tech job

| In Career & Growth
| 6 minute read

A YouTube viewer reached out to me the other day after watching one of my videos. He’d just started at a big tech company and wanted to know: How do you navigate the fog? How do you figure out what to focus on? What are things to avoid?

I’ve felt that fog three times—at Sun Microsystems, Oracle, and AWS. And after 27 years and countless conversations with mentees, I think there are three phases that help cut through it.

A tabby cat with striking amber eyes sits at a desk in warm, golden lighting, surrounded by an open book, stacked papers, and a glowing desk lamp. Digital icons representing WiFi, images, shopping, and other technology symbols float in the air around the cat, suggesting connectivity and digital communication. The scene is bathed in warm orange and yellow tones, with a window visible in the background showing an evening sky.

We’re still living in the cat pictures era of AI

| In AI
| 7 minute read

Most people use AI the way our parents used the Internet in 1995. They’d dial up their modem, check the weather forecast on Yahoo, maybe look up a recipe, and call it a day. The idea that this same technology would eventually let them video call their grandchildren, run a business, or access humani…

AWS re:Invent 2014 conference slide showing Constantin in a light blue shirt standing on stage. The slide title reads “What you’ll get out of this session” with five bullet points: A lower AWS bill, A more scalable, robust, dynamic architecture, More time to innovate, Real-world customer examples, and Easy to implement. A mountain with a flag icon appears in the bottom right corner of the slide.

3 slides everyone uses (but you should delete)

| In Career & Growth
| 7 minute read

In 2014, I was preparing my AWS re:Invent presentation on “Running Lean Architectures.” I had my slides ready: agenda, bio, resources. Then I stopped. Why was I presenting this in the first place? The answer: to help people save money on AWS. Not to check boxes on a corporate template. Not to prove …

A watercolor illustration of a solitary figure in dark clothing standing on a vast, layered desert landscape. The scene features rolling sand dunes in warm tones of beige, tan, and gold, with deep burgundy and rust-colored bands flowing across the foreground. A golden sun or citrus slice appears in the upper left corner. The composition conveys a sense of isolation and contemplation in an arid, minimalist environment.

When Claude Suggested Brian Eno: Building Art You Can’t Control

| In AI
| 11 minute read

On constrained serendipity, learning by doing, and whether the system is the art. I’m chatting with Claude, kicking around ideas on building a tool that generates AI images, but without the prompting. I like to start these chats by brain-dumping ideas into it. In this case it’s about auto-generating…

A blue background with wavy patterns features the text ‘INVITE YOUR HEROES INTO AI!’ in white letters with red outline. On the left is a person in a purple suit with a dancing pose. Next to them is a large yellow pointing hand gesture towards a robot with a square head, round eyes, a red triangular nose, and a yellow bulb on top. On the right is a cartoon illustration of Constantin, wearing a dark green hoodie, with a thoughtful hand gesture near his chin.

Invite your heroes into your AI conversations

| In AI
| 7 minute read

I had a problem.

I wanted to create a YouTube channel and had my first video ready to go, but then… nothing. Weeks passed. I had ideas, I had equipment, and I’d watched enough tutorials to edit like a pro. But I couldn’t get myself to shoot video number two.

Classic procrastination.

So I did what I often do when I’m stuck: I talked to Claude. But this time, I didn’t just ask for advice. I invited two experts into the conversation to help me figure this out.