iiinova — where your digital self keeps evolving

Georg Unterholzner @georg.dev · 23/02/2026

When I first saw it working, me reviewing from my phone while the AI iterated on my feedback, I understood this is the way forward. We don't need to reinvent the wheel. We have decades worth of tools and practices. We just need to let AI meet us where we already are.

110

Georg Unterholzner @georg.dev · 23/02/2026

kipppunkt/build fixes that. You review along the way, leave feedback, change requirements on the fly, all directly from the PR. As a bonus, it lets you run multiple agents in parallel with dependency management and automatic merge conflict resolution.

100

Georg Unterholzner @georg.dev · 23/02/2026

"Isn't that just a ralph loop?" Kinda, but not really. Ralph loops are basically waterfall. You hit go and hope for the best and get no feedback until it's done. They fail the same way waterfall always has.

120

Georg Unterholzner @georg.dev · 23/02/2026

It picks up your requirements, implements them, opens PRs on GitHub, and responds to your review comments. You review from wherever. The couch, a café, the beach. You don't need a TUI or pair-programming with a chatbot. Just requirements in, PRs out.

100

Georg Unterholzner @georg.dev · 23/02/2026

Here's a glimpse of how software engineering is going to look soon. I wanted to spend more time with my newborn daughter and less time sitting in a terminal watching an AI type. So I built kipppunkt/build. Today I'm sharing it. github.com/kipppunkt/bu...

120

Georg Unterholzner @georg.dev · 02/02/2026

If you're building TUIs with Claude Code or similar AI agents, I made an agent skill for Ink. It organizes the docs so the agent can pull in only what it needs on demand, saving you lots of tokens.

github.com

GitHub - georg-unterholzner/ink-skill: Agent Skill for building CLI applications with Ink

Agent Skill for building CLI applications with Ink - georg-unterholzner/ink-skill

000

Georg Unterholzner @georg.dev · 09/01/2026

Five-minute timelapse shot on a Pixel 9 on my last day of vacation in South Tyrol, Italy. At some point in my life, I'll move back here, get myself the best telescope money can buy, and spend my evenings taking photos of galaxies and nebulas far away.

030

Georg Unterholzner @georg.dev · 16/12/2025

I interviewed someone for a dev position the other day. Remotely. Asked the candidate a very vague question. Got an extremely thorough answer. Really impressive. Left a weird aftertaste, though. I just typed the question into ChatGPT. Same answer 💀

000

Georg Unterholzner @georg.dev · 10/12/2025

Tbf, the bug was very subtle which is also why it took so long to be found. But you're right, the whole feature seems like a high risk, low reward situation. It's cool tech but nothing I'd want to use in prod.

010

Georg Unterholzner @georg.dev · 09/12/2025

It's kinda crazy how React went from "it's just a super lightweight rendering library, bro" to "whoopsie, our server-side execution layer's complex custom data-format serializer had a prototype-pollution bug and now it's a CVE-10 RCE."

100

Georg Unterholzner @georg.dev · 03/12/2025

I put everything into this article including the link to an example repository and the full prompt file, so that you can just copy/paste it into your repo:

georg.dev

AI-Driven Unit Testing with GitHub Copilot • georg.dev

A better alternative than vibe-coding unit tests

000

Georg Unterholzner @georg.dev · 03/12/2025

It uses many of GitHub Copilot's capabilities (instruction files, prompt files, some tools & configs) for prompt and context engineering. With that, it's reliably able to one-shot entire test suites in my day-to-day work. See for yourself (select 2x speed):

100

Georg Unterholzner @georg.dev · 03/12/2025

1. The AI figures out business logic & edge cases and then proposes tests based on that. 2. You approve or modify the list. 3. Then Copilot implements everything autonomously while you do whatever. It's a little bit more complicated than that but that's the gist of it.

100

Georg Unterholzner @georg.dev · 03/12/2025

AI-generated unit tests suck. Claude overshoots. GPT-5 is lazy af. No model has proper common sense. I tinkered around for a long time to get GitHub Copilot to actually work as expected. The result is this workflow.

110

Georg Unterholzner @georg.dev · 18/11/2025

I wanted to announce a new blog article today. My blog runs on @cloudflare.social Pages. Good that I didn't.

020

Georg Unterholzner @georg.dev · 17/11/2025

Think you're safe because you're in a metro station? Think again! Little Berta up there is watching you. Luckily, she’s nice enough to leave hints on the floor to spare those who paid attention in Stats 101.

000

Georg Unterholzner @georg.dev · 17/11/2025

You'd be mistaken if you thought pigeon droppings on the floor were just a random annoyance. Quite the opposite! They're a real-time heatmap showing you the probability of being hit by pigeon poop at any given spot. And you thought all that statistics knowledge was useless!

110

Georg Unterholzner @georg.dev · 13/11/2025

One killer use case for AI agents is making sense of minified JavaScript files. I just built a post-install script that monkey-patches a package I depend on. What would have easily taken me half a day was finished in under an hour.

000

Georg Unterholzner @georg.dev · 12/11/2025

Very weird and I'm kinda nostalgic for RxJS. But seeing an `effect()` inside a constructor is something I probably still have to get used to.

000

Georg Unterholzner @georg.dev · 12/11/2025

Turns out my team and I are partially doing Angular ... again. Haven't touched the framework since v16. Feels like ages.

110

Reposted by Georg Unterholzner

Sarah Drasner @sarahedo.bsky.social · 06/11/2025

🛠️ If your team has issues to work through: We do an exercise at offsites called Elephants, Tigers & Paper Tigers. - Elephants are things that the group isn’t talking about but needs to - Tigers are things threatening the team, risks - Paper tigers are things that seem like risks, but aren’t

815926

Georg Unterholzner @georg.dev · 20/08/2025

Sounds very interesting! I have a lot on my plate at the moment as well, but I could spare a few hours per week for a promising project. Is there a GitHub repo or something similar?

000

Georg Unterholzner @georg.dev · 20/08/2025

Sounds awesome! Where can I sign up?

110

Georg Unterholzner @georg.dev · 20/08/2025

Oh boy, you're not talking about running different frameworks in iframes and plugging them together, but about running multiple frameworks in a single browsing context?

110

Georg Unterholzner @georg.dev · 20/08/2025

That's the most interesting part! The state needs to be serializable, though.

100

Georg Unterholzner @georg.dev · 20/08/2025

Is there a good way to test at least some of those requirements automatically in CI? I only know of Axe but that won't test for screen reader interoperability for example.

100

Georg Unterholzner @georg.dev · 19/08/2025

Getting a full afternoon of uninterrupted coding is so nice. I don't get that often but today is one of those days. Feels great!

000

Georg Unterholzner @georg.dev · 18/08/2025

Sometimes programming just feels too much like Hänsel and Gretel.

const breadcrumbsForChildren = generateBreadcrumbs(children);

020

Georg Unterholzner @georg.dev · 13/08/2025

I found people calling the second one "children pattern". This concept isn’t React specific, though, probably not even JavaScript specific. There must be framework & language agnostic terms. But which? - config vs. extension point pattern? - props vs. slots pattern? - ?

000

Georg Unterholzner @georg.dev · 13/08/2025

Basic question but I can't figure it out. How are these two design patterns called? In the first, the component structure is fixed, and behavior is controlled *only* through props serving as config. In the second, an extension point via the children prop allows for more flexibility.

Component structure is fixed, and behavior is controlled *only* through props serving as config.

Extension point via the children prop allows for more flexibility.

100

Georg Unterholzner @georg.dev · 13/08/2025

Copilot: "I cannot work properly without an instruction file." Also Copilot: "Just let me write the instruction file myself."

000

Georg Unterholzner @georg.dev · 01/08/2025

Don't be like me. Don't sleep on @playwright.dev's test steps. It took me 4 years to realize how much cleaner, readable, and debuggable my tests could be. Named steps = better readability, clearer reporting, faster debugging. Check out the before/after!

020

Georg Unterholzner @georg.dev · 30/07/2025

Some tools force you to learn a complicated API and require constant maintenance. Other tools... just work, and that's why it's easy to overlook them. @openfeature.dev is one of those tools. Do I use it daily? Yes. Do I ever have to think about it? No.

022

Georg Unterholzner @georg.dev · 28/07/2025

Finally moving all my pet projects off Travis CI. Not a fan of their expiring credits billing model. But where to migrate? - Circle CI - GitHub Actions - ...? What’s your pick?

000

Georg Unterholzner @georg.dev · 22/07/2025

Want step-by-step guidance + templates to create your own copilot_instructions.md? I wrote a detailed guide to help you get started: georg.dev/blog/05-bett...

georg.dev

Better Context, Better GitHub Copilot | georg.dev

A guide to copilot_instructions.md

000

Georg Unterholzner @georg.dev · 22/07/2025

Agent mode *bonus* tip: Enable auto-approval for tools by adding "chat.tools.autoApprove": true to your settings.json. Just be aware that you’re giving Copilot a free pass. Use it wisely! 😅

110

Georg Unterholzner @georg.dev · 22/07/2025

Agent mode tip: Add clear instructions for automating checks like "npm run lint" or "npm run test". With these in your file, Copilot Agent can run the commands, fix errors, and iterate until all checks pass without constant back-and-forth.

Screenshot from VSCode showing GitHub Copilot Agent asking for approval to run the lint command

100

Georg Unterholzner @georg.dev · 22/07/2025

Most advice skips this: Be model-specific. If Copilot repeatedly makes the same mistakes (e.g., overly verbose code or missing edge cases), include correctional hints in your file. Example: "Focus on simplicity, avoid over-comprehensive solutions."

100

Georg Unterholzner @georg.dev · 22/07/2025

Some essential sections to include: - Summary: What the project does, briefly. - Terminology: Domain-specific terms Copilot can’t infer. - Architecture: Key files and why things are structured this way. - Task planning: Steps Copilot should follow before coding.

100

Georg Unterholzner @georg.dev · 22/07/2025

Most devs either skip copilot_instructions.md or fill it with generic fluff. But irrelevant instructions waste Copilot’s context space and reduce output quality. Focus on non-obvious details: unique constraints, domain rules, and project specifics.

100

Georg Unterholzner @georg.dev · 22/07/2025

I was frustrated with the lack of clear advice on setting up copilot_instructions.md. Most guides are vague, or worse, generic AI slop that *hurts* Copilot's performance. So I experimented and found better ways to make Copilot *actually helpful*. 🧵

122

Georg Unterholzner @georg.dev · 17/07/2025

Stack Overflow, are you alright?

000

Georg Unterholzner @georg.dev · 16/07/2025

Absolutely. It feels like we spent decades talking about 'deep work' and 'flow state' only to abandon those concepts the moment AI companions showed up.

010

Georg Unterholzner @georg.dev · 14/07/2025

Finally!! Grok 4 in GitHub Copilot.

010

Georg Unterholzner @georg.dev · 28/06/2025

Gen Alpha successfully rotted my brain. grid-template-columns: 1fr 2fr; now reads “one for real, two for realz.” My grid is aggressively verified. ✅

Post on X from @matthewcp:
"Is it? display: grid, grid-template-columns: 1fr 2fr; or whatever, that's what you need for typical flexbox usage."

100

Georg Unterholzner @georg.dev · 23/06/2025

I was looking for exactly this for months. Thank you for sharing!

010

Georg Unterholzner @georg.dev · 23/06/2025

Crazy idea to utilize the webcam input. Mind-blowing.

010