Automated tasks need to be monitored in case something is off: a github workflow that triggers a different workflow via curl received HTTP status 204 for about 10 days in a row. Then it returned 200 instead. The updated tasks accepts all 2xx codes. Problem solved. #MercilessAutomation #Engineering
I found myself typing "git pull" followed by "code ."
To make my life easier I introduced an alias "gpc" that executes these two in one step.
It may sound like a small thing but a large number of small improvements will make a noticeable difference.
#Engineering #MercilessAutomation
Today, Github Copilot is shockingly slow. Plenty of time to grab a coffee ... ๐ฑ
Microsoft Outlook? Continues to suck.
Newest example: I have an existing contact with an email address. Then I need to change the email address. When I try to add that contact to an invite, it continues to show the old email address.
Why?
#FunWithMicrosoft #Outlook #Microsoft
You can use Flutter to build some pretty sleek, responsive apps. But to get the most out of it, you'll need to understand async programming concepts like streams. Here, Anthony explains what streams are and how they work in Flutter, including real-world examples.
www.freecodecamp.org/news/how-to-...
The correct version is VS Code 1.104.x or later. Apologies for the typo. ๐ #VSCode #DevContainer
I also respond best to poems
A silent Sunday to me means uninterrupted time to look into one of the more challenging problems I encounter as a software craftsman. Without pressure and the environment calm and quiet, I find problem solving is much easier. #SilentSunday
When you upgrade to VS Code 1.140.x or later, and you're using dev containers, it may fail starting the container when it tries to download the VS Code Server. In that case uninstalling and reinstalling the "Remote Development" extension resolved the issue in my environment. #VSCode #DevContainer
Big question as often: how many minutes to I wait for Github Copilot using GPT-5 (preview) before I call it quits? 1 min, 2 mins, 5 mins ... Tough call! ๐ #FunWithAI #LLMReality
A watch showing the time 10:09, either am or pm.
"Create an an image of an analog watch showing the time 1:25 pm". This is what ChatGPT 5 created. #FunWithAI #LLM #AI
We have: ``` 5.9 = x + 5.11 ``` To find `x`, subtract `5.11` from both sides: ``` x = 5.9 โ 5.11 ``` Step-by-step: 1. Convert both to hundredths for clarity: ``` 5.90 โ 5.11 ``` 2. Subtract: ``` 5.90 ``` * 5.11 --- 0.79 ``` So: ``` x = -0.21 ``` โ **Answer:** **x = -0.21** ```
Impressive PhD-level math skills demonstrated by ChatGPT 5. Have a look and enjoy!!
Honestly: I think the statement that ChatGPT 5 is at "PhD-level" is an insult to all people with a PhD.
#FunWithAI #ChatGPT #LLM
A source code snippet with Github Copilot suggesting a random import statement at the end of a TypeScript file. Note that imports are typically at the very beginning of the file. I have all imports at the beginning of my TypeScript files, so Copilot can't claim it doesn't "know".
Github Copilot getting bored and starts making random suggestions that make no sense, see the last line in this screenshot.
I think Copilot has mood swings as well. On some days it works quite well when you spoon feed it. On other days, e.g. today, the error rate is closer to 80%.
#FunWithAI #LLM
I've tried ChatGPT 5 in the last few days. Today speed is atrociously slow. I can easily get a coffee between replies. I'm not noticing a quality improvement yet. Given that, I just switched back to ChatGPT 4o, and the response times are much more reasonable.
#FunWithAI #AI #ChatGPT #LLM
public async process(event: IDomainEvent): Promise<void> { switch (event.eventType) { case 'account.created': case 'account.updated': case 'account.deleted': console.log(`${timestamp()}: loc 250731-2002: Processing event of type ${event.eventType} for entity OID ${event.entityOid}`); break; default: // Ignore unknown event types, do not log or call any handler return; } switch (event.eventType) { case 'account.created': await this.handleCreated(event); break; case 'account.updated': await this.handleUpdated(event); break; case 'account.deleted': await this.handleDeleted(event); break; } }
This is the "excellent" code quality produced by an AI agent. A human coder would have expressed this in a single switch()-statement which would also be easier to read and understand.
Looking at this, I'm wondering how bad is the training data really?
#FunWithAI #llm #Agents
The picture shows the graph of the Microsoft stock on 25 July 2025. In after hours trading it contains a few spikes that appear to be low-ball offers to buy or sell. As a result the part without the spikes has become virtually unusable.
An example of a graph that does show the actual quotes at the time. However, the graph isn't really useful this way. (Source: finance.yahoo.com, as of 25 Jul 2025 at end of after-hours trading). One option to improve it would be an optional filter to catch spikes, up and down. #BugCollection
What a useful message! ๐
(Source: Left feedback on a VS Code Documentation page.)
AI in the form of LLM's and it their present state that are intended to generate code or assist with writing code allows me to produce garbage code at an unprecedented rate: I've become a 10x developer!! ๐
An incredible amount of editing is still required.
#TheRealityOfAI
Don't we all just enjoy fixing the bugs that were introduced by AI because it suggested code that looked correct. And yet, some subtle difference actually turned out to be a hallucination. Going through a piece of code with a range of bugs created by AI. What a nightmare! #TheRealityOfAI
I just removed 5 empty lines in a file to improve the code layout. Zero functional changes. Commit message suggestion by Copilot: "refactor AccountCommandService to use saveAccount method for persisting domain events and index accounts with SearchServiceAdapter" - #TheRealityOfAI
New LTS version of Node.js is available: 22.15.0. Download from here: nodejs.org/en/
Also note that Node.js v18 has reached end of life on 30 April 2025.
#Nodejs
Tab programming at its finest. AI in a death spiral ..... I'm observing this multiple times a week. Sad! #TheRealityOfAI
How stupid: suggesting a method that is already there, even if it's the minimum. The AI should be aware there is a method of that name already and at best suggest changing it's implementation. #TheRealityOfAI
Awesome! I have 6 VS Code instances open with one dev container each plus dependencies. Then CoPilot chart stops responding. A few minutes later all VS Code instances crash.
I wish they got the quality of the base product right before releasing yet more broken features ...
#TheRealiltyOfAI
AI getting into the way with TDD: AI writes a test for the existing (minimal!) implementation which just throws an exception. This test would cast this in concrete instead of driving the correct implementation. This demonstrates that the AI doesn't "understand" the concept of TDD. #TheRealityOfAI
After Microsoft, Amazon now also adjusts at least some of its capacity plans for new data centers. Microsoft referred to lower demand than anticipated.
finance.yahoo.com/news/amazon-...
#TheRealiltyOfAI #AMZN #MSFT
The "loop of death" increasingly becomes a common theme. Typically observed on interfaces, it also struggles to generate correct code for a constructor, despite that constructor being in the code based and imported. It just keeps making stuff up. (Copilot / GPT-4o / Agent Mode). #TheRealityOfAI
Only one way to make a worse suggestion: A function that generates purely random tokens. Here it suggests a non-existing identifier named "mockAlpaca" ... I can add it but expected the generated code to be correct in the first place..
AI forces me to clean up after it ... again.
#TheRealityOfAI
