Remote

ViennaMike

@ViennaMike@mastodon.social

mastodon 4.8.0-nightly.2026-10-06

Computer and electronics hobbyist. Intelligent Transportation Systems consultant. He / Him

43 Followers

79 Following

22 Posts

Joined November 09, 2022

website:

https://www.mcgurrin.info/robots/

website:

https://www.mcgurrin.info

GitHub:

https://github.com/ViennaMike

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@cstross@wandering.shop So the experience of developers who have found significant value from the use of LLM's doesn't count or isn't valid? Used to WRITE material, they aren't great. But to organize, summarize, do research and help brainstorm, they can be useful. While ChatGPT currently can't pass the IRS' volunteer tax assessor's exam (see https://www.mcgurrin.info/robots/8298/) I have confidence that a RAG based system could do so. Heck, COMPUTERS are magic to most people, as are the internet, radios and TV.

Can ChatGPT Certify as a Tax Preparer? | The Aspiring Roboticist

3

4

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

One of the benefits of AI assisted development that I'm sure many practitioners understand but I don't see talked about much, is the ease of keeping documentation up to date. After making additions and/or changes, you can just ask the AI to update the documentation to capture the changes. As a human, you can and should review and verify the proposed changes.

It's an incredible time saver AND makes if far more likely you WILL keep the documentation up to date.
#AI #Claudecode

mastodon.social

Mastodon

1

2

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

Home use for AI: Our user manuals, warranty cards, etc. were totally unorganized and in several locations. As I'm grouping them roughly alphabetically, I'm snapping a picture of the cover of each. Claude scans the photos and generate a spreadsheet with the category, brand, model or product, Item type, and documentation type. It pulled it all from the photos and created a spreadsheet, no typing required.

You could do the same for a record collection, book collection, or anything similar.

1

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@cstross@wandering.shop Love your fiction, can't agree on this take. Spam does not help with writing code, whereas LLM's can be extremely helpful. LLM's are not generally intelligent. They can be immensely problematic, but to dismiss them as merely spam generators as you have, or "parrots" as others have is simply incorrect.

2

20

0

0

ViennaMike @ViennaMike@mastodon.social

· 9mo ago

Replying to

@cstross@wandering.shop @jackwilliambell@rustedneuron.com @ravenonthill@mastodon.social Back in the day, at least one model of X.25 packet switches from Hughes were programmed in Pascal. This was when Pascal was the big teaching language. As it was explained to me when I asked "why not C?", Pascal was fast enough to meet the requirements and they could hire new grads who knew Pascal without having to train them in C. The FAA's NADIN II packet switching network (long since retired) used these switches.

2

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 6mo ago

Building My First RAG System (Which Sadly Failed the Tax Preparer Exam)
A while back, I posted about how I gave the IRS’ volunteer tax preparer exam to ChatGPT. It scored in the 70s, but an 80% is the minimum required score to pass and be certified. A short while later, I took a Coursera short course on Retrieval Augmented Generation (RAG). I then decided to
https://www.mcgurrin.info/robots/8318/
#Uncategorized #AI #chatbots #chromadB #gradio #llamaparse #LLMs #ollama #python #RAG #Taxes

1

0

1

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@VictimOfSimony@infosec.exchange I'd been following him for some time, because yes, I agree that he is entertaining. And I'm a fan of his Laundry File fiction.

1

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@leadegroot@bne.social @dragonfrog@mastodon.sdf.org @ThePowerNap@mefi.social I can certainly see a bubble bursting, but I'm not sure that Phind is a sign. Hard to compete with those with billions to develop core models, and not going to hit any profit home runs reselling access to other's models.

1

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@davidgerard@circumstances.run I read both the article and the referenced paper that you cited. The paper that I provided DOES use pull requests as the key metric, but looks at others as well, including code quality. The Stanford study that I provided explicitly does NOT use pull requests as the productivity metric. I take it you didn't watch the video? And the Stanford study is far larger and more comprehensive than the one you cited.

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

Hauntomation: Automating Halloween with Python
PyCon US 2026 — Lightning Talk Companion Page

If you scanned a QR code from a five-minute talk and ended up here: welcome. Everything I showed and links to additional information is here.

If you didn’t see the talk and just wandered in: I build automated Halloween props as a hobby, often using Python.
The talk

Slides (PDF): [link to be added after the talk]
Recording: [link to be ad
https://www.mcgurrin.info/robots/pycon2026/

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

I'm using Claude Code on a project. As it was thinking about how to code the next piece and giving status updates, it reported, "Consulting the rubber duck." 😂 Those who know will know!

#AI #rubberduck

mastodon.social

Mastodon

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

Pure "vibe coding" might work, or it might lead to disaster. But working with AI to develop architecture / requirements / design notes before coding? Speed and quality. #AI

mastodon.social

Mastodon

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

Virginia has just become the 19th state to join the National Popular Vote Interstate Compact. Can you help get your state to join?
https://myemail.constantcontact.com/Virginia-becomes-19th-state-to-enact-National-Popular-Vote.html?soid=1116971938310&aid=cKiBOjnxjt0

myemail.constantcontact.com

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

As we celebrate the successful Artemis 2 mission, today also marks the 45th anniversary of STS-1, the first Space Shuttle mission. To-date, it's the only launch I've personally witnessed live.

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 5mo ago

Quick Post #7: AI for Cataloging in the Home
Home use for AI: Our user manuals, warranty cards, directions, etc. for electronics, appliances, etc. were totally unorganized and in several locations. As I'm getting them out and grouping them roughly alphabetically (folders for every few letters), I'm snapping a picture of the cover of each one. Claude is then able to scan the photos and generate a spreadshe
https://www.mcgurrin.info/robots/8330/
#Uncategorized #AI #cataloging #Claude #LLMs

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@davidgerard@circumstances.run @cstross@wandering.shop The article you posted included a very useful study. However, contrary to the article, that study is NOT the only study on the topic. Other, far larger, more comprehensive, studies show measured, meaningful productivity improvements. You asked for measurements, so here they are: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4945566 and https://www.youtube.com/watch?v=tbDDYKRFjhk The Stanford study found that the average productivity boost is significant (~20%) but some teams see productivity decrease.

0

2

0

0

ViennaMike @ViennaMike@mastodon.social

· 45mo ago

Replying to

@dan@discuss.systems Doesn't that depend on the protocol?

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@VictimOfSimony@infosec.exchange @cstross@wandering.shop If you want to debate whether or not, as a whole, LLMs are a net positive or net negative, I agree. Intelligent people can legitimately disagree. One cannot, based on the evidence, legitimately claim that they only produce spam.

0

4

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@VictimOfSimony@infosec.exchange Well, I won't know what @cstross thinks, as he apparently has blocked me from pushing back on his hyperbole.

0

2

0

0

ViennaMike @ViennaMike@mastodon.social

· 6mo ago

Replying to

@mcc@mastodon.social That's why I support the Virginia initiatives for Ranked Choice voting: not perfect, but a marked improvement, and we're actually making progress on getting it more broadly adopted. It's instant runoff for single.positiom and Single Transferable Vote for multiple seats.

0

0

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@ThePowerNap@mefi.social @cstross@wandering.shop I don't disagree. I think part of the path going forward will be to push more to the edge, as we've seen with other types of AI, using less powerful, but good enough, models for many applications. Models that don't require as much energy to run.

0

2

0

0

ViennaMike @ViennaMike@mastodon.social

· 8mo ago

Replying to

@mrjunge@raru.re If you have to spend almost as much time, or more, cleaning up an LLM-assisted product as you originally saved, I agree. LLM's can certainly generate spam or spam-like output at times. But to claim that that's ALL it does is not correct. One company found that initially, their software code writing gains were eaten up by quality control rework, but that after learning how to better use the tools, when to use them, and when not to use them, production went up 30% with no loss of quality.

0

0

0

0