> where we identify public servants with strong technical aptitude across government, bring them into dedicated product teams
> The team’s approach was straightforward. Build working software fast. Put it in front of real users early. Collect feedback. Fix things quickly. Release updates every two weeks.
> That’s a 95% cost reduction. Both systems instead of one. Delivered faster. With 643 users already on the platform
This is a proven solution. These parts, the non-AI management ones, are proven to work in all sorts of places. Gov.uk is another example.
However, there's one massive problem with this: it doesn't involve the free market and it doesn't make any money for corporations to feed back to politicians in campaign donation kickbacks. It even involves respecting civil servants - maybe even paying them market wages! These parts are so heretical that most governments would choose the solution that 10X more expensive and also doesn't work, every single time.
Great idea, and hopefully great results. But it’s written like LinkedIn “broetry” and that AI image at the top promises a fluffy article. Maybe expand a bit on some of the impressive tech described in the body?
I’m at the point where if I see an AI image at the start of an article I just back right out. It would be so much better if the author just didn’t include an image at all. What did this image actually add to the content of the post? If you’re just doing something for the sake of doing it you’ve lost the plot.
I've been saying that for a while now. They could just have put an Alberta landscape shot or something else that's actually aesthetically pleasing and relevant, and doesn't feel as trashy as the picture actually in the article.
I genuinely don't understand this trend of slapping AI slop on top of your article. I've even seen good articles do it.
An image signals that the author put time and energy into the article and that they have an eye for detail. Even if it's an AI generated image because the author still had to pick a fitting image.
I disagree. Especially with AI, it’s far too easy to generate and insert an image with no time, energy, or eye for detail.
Authors do it because it supposedly leads to better engagement, shows up bigger on social media, and breaks up the text. But generally, unless the visual content meaningfully adds to the text content, users will largely ignore it.
If you generate an image that contains a bunch of low-quality gobbledegook, the signal that it sends is that you have zero eye for detail. Look at any single part of that image, and you will notice how little detail there actually is.
The tech isn't the big news. It's the process opportunity for governments. That said, I did really like what they reported on a use case of Gemini -- they got a bunch of people to video every single process in the old systems, and then got Gemini to watch the video and write full specs for the new systems. Niiice.
What you have in civil government is a lot of people and a lot of time -- turning that into inputs for acceptance on the new codebase is super smart -- and using only their expertise (legacy system screen caps), but relying on the AI to do all the tech spec work feels super smart.
This sounds like the exact type of system my team replaces. 4 years and $54M sounds like some consultancies saw a money piñata (government contract) and decided to take a crack at it.
There is no reality on this planet where this project costs $54M and 4 years. Good on the folks in the ministry to notice this and not blindly follow implement.
"On the client side, this project would not have been possible without the exceptional support of Deputy Minister Mark Kleefeld and his team at the Ministry of Infrastructure. From the very beginning, Infrastructure’s leadership understood what we were trying to accomplish and backed it fully. That kind of top-down support from the client ministry is rare, and it made all the difference."
That's kind of amazing. Alberta has a conservative govt so I am surprised "in-house" got the pass over "outside company". It is good to see fiscal conservatism over 'govt-bad' conservatism. Hats off to the deputy minister et al. for approving this.
Using Google Gemini to generate requirements/spec document from video is amazing. I wonder what the prompt looked like and if there was custom support to help process the videos.
Deputy ministers are supposed to be the top non-political/bureaucrat position in a ministry, so theoretically it wouldn't matter if the government is conservative. The elected minister sets the goals and gives general directives, the professional deputy is in charge of execution. In reality this neat distinction is often broken, some governments politicize appointments even below the minister level, etc.
Writing quality was distracting. Very breathey. Hard to understand if I was getting important information or not -- but it's ok, some people will defend this style.
You can look at a histogram of number of words per sentence, and you'll find immediately that it's written by an AI.
When?
Today.
Minutes.
Four years.
$54 million.
Collect feedback.
Delivered faster.
Not days.
Not weeks.
It's free.
...
No $19 million in upfront costs.
They're now doing meaningful work.
Let me put that in context.
That's a 95% cost reduction.
But think about what that represents.
And we can show you how.
The only time I want to read something written by AI is when I am interacting with an LLM directly. I don’t know who these people are that think others want to read their slop. Even if the content is good, I can’t stand reading it.
> They understand the business processes they’re digitizing.
I feel this has more importance than they think. Outside consultants would not have had this domain knowledge and would have spent months learning it. And then would have had to fix their mistakes because they misunderstood something (billed to the province, naturally)
We're having the users who understand the business processes use Claude to create clickable demos. Then these are presented to the engineering team to re-build. The users are loving it because they can ask for exactly what they want.
The shortlist came down to four major consulting firms.
This is the problem in a nutshell. Those firms are structured to extract money from their customers, not to produce useful work. The fact that anyone is signing contracts with them any more blows my mind.
I wonder how much of the needs satisfied by this software in Alberta is also needed by other states?
Yes, it's great that they saved money and built what they need so quickly, but a catch with govt is that they build out these proprietary processes that need more bespoke software.
Alberta needs to leave Canada. Imagine having the profit from all of your natural resources leave the state to fund social programs that don't benefit you at all. The rest of Canada does nothing for Alberta.
I think the use of AI is really missing the point here. The point is that small in-house teams can deliver a lot more quickly and to a higher quality and at a lower cost than large outsourced teams from the big consultancy companies. I've seen this over and over again (the problem is that large organisations often prefer to go the slow and expensive route with the big consultancy companies for a complex variety of reasons). So it would be like an article saying "our small inhouse team using VS Code did a much better job than a big outsourced consultancy using MS Visual Studio - isn't VS Code awesome".
I'm using AI every day (and it's not really about AI), but:
Anyone else closed the article immediately after seeing the low-taste, sloppy image at the top?
How do you call this aesthetic? "Futuristic vomit"? AKA "Generate image of: code blocks, AI-brain image, diagram, smiling guy and bunch of other crap. Make it look cool and futuristic, make no mistakes"?
Anecdata: while i was in a tiny tiny software company, we got an in at a large auto manufacturer. They said they had been trying to get someone to do that job for like 2 years.
The job was of the 'two people 3 months' magnitude. The procurement system was also of the 'two people 3 months' magnitude so we simply gave up.
In the article's case, they could have done this even before coding assistants. It would have cost the estimated 5 million instead of 850k, but that's still 10x less than the 54 million.
Yes, I think the innovation here is social as much as technical. The AI Magic Sprinkle fairy dust on the proposal to solve internally, is what grants permission to go ahead. It both increases the chances of actual success and gives something external to put the blame on if something fails.
Mm, did you read the part where they have hundreds of people in and reporting bugs and feature requests every week? Testing is probably too granular a concept for the target audience of this blog post.
> where we identify public servants with strong technical aptitude across government, bring them into dedicated product teams
> The team’s approach was straightforward. Build working software fast. Put it in front of real users early. Collect feedback. Fix things quickly. Release updates every two weeks.
> That’s a 95% cost reduction. Both systems instead of one. Delivered faster. With 643 users already on the platform
This is a proven solution. These parts, the non-AI management ones, are proven to work in all sorts of places. Gov.uk is another example.
However, there's one massive problem with this: it doesn't involve the free market and it doesn't make any money for corporations to feed back to politicians in campaign donation kickbacks. It even involves respecting civil servants - maybe even paying them market wages! These parts are so heretical that most governments would choose the solution that 10X more expensive and also doesn't work, every single time.
Great idea, and hopefully great results. But it’s written like LinkedIn “broetry” and that AI image at the top promises a fluffy article. Maybe expand a bit on some of the impressive tech described in the body?
I’m at the point where if I see an AI image at the start of an article I just back right out. It would be so much better if the author just didn’t include an image at all. What did this image actually add to the content of the post? If you’re just doing something for the sake of doing it you’ve lost the plot.
I've been saying that for a while now. They could just have put an Alberta landscape shot or something else that's actually aesthetically pleasing and relevant, and doesn't feel as trashy as the picture actually in the article.
I genuinely don't understand this trend of slapping AI slop on top of your article. I've even seen good articles do it.
An image signals that the author put time and energy into the article and that they have an eye for detail. Even if it's an AI generated image because the author still had to pick a fitting image.
I disagree. Especially with AI, it’s far too easy to generate and insert an image with no time, energy, or eye for detail.
Authors do it because it supposedly leads to better engagement, shows up bigger on social media, and breaks up the text. But generally, unless the visual content meaningfully adds to the text content, users will largely ignore it.
Two posters above found that the signal was more like “get ready for a gen-AI article filled with vague woolly sentences”.
Wasn't there anything relevant available? Screenshots of the new tools in a before/after collage perhaps?
An AI image signals that the author did not put time and energy of their own in, they had the AI make it up. It's a yellow flag.
If you generate an image that contains a bunch of low-quality gobbledegook, the signal that it sends is that you have zero eye for detail. Look at any single part of that image, and you will notice how little detail there actually is.
If that's the image they picked, I question their taste.
The tech isn't the big news. It's the process opportunity for governments. That said, I did really like what they reported on a use case of Gemini -- they got a bunch of people to video every single process in the old systems, and then got Gemini to watch the video and write full specs for the new systems. Niiice.
What you have in civil government is a lot of people and a lot of time -- turning that into inputs for acceptance on the new codebase is super smart -- and using only their expertise (legacy system screen caps), but relying on the AI to do all the tech spec work feels super smart.
Totally. A small motivated team working iteratively with the users of the current system. That’s a great way to work, regardless of the AI “hook.”
This sounds like the exact type of system my team replaces. 4 years and $54M sounds like some consultancies saw a money piñata (government contract) and decided to take a crack at it.
There is no reality on this planet where this project costs $54M and 4 years. Good on the folks in the ministry to notice this and not blindly follow implement.
"On the client side, this project would not have been possible without the exceptional support of Deputy Minister Mark Kleefeld and his team at the Ministry of Infrastructure. From the very beginning, Infrastructure’s leadership understood what we were trying to accomplish and backed it fully. That kind of top-down support from the client ministry is rare, and it made all the difference."
That's kind of amazing. Alberta has a conservative govt so I am surprised "in-house" got the pass over "outside company". It is good to see fiscal conservatism over 'govt-bad' conservatism. Hats off to the deputy minister et al. for approving this.
Using Google Gemini to generate requirements/spec document from video is amazing. I wonder what the prompt looked like and if there was custom support to help process the videos.
Deputy ministers are supposed to be the top non-political/bureaucrat position in a ministry, so theoretically it wouldn't matter if the government is conservative. The elected minister sets the goals and gives general directives, the professional deputy is in charge of execution. In reality this neat distinction is often broken, some governments politicize appointments even below the minister level, etc.
Writing quality was distracting. Very breathey. Hard to understand if I was getting important information or not -- but it's ok, some people will defend this style.
You can look at a histogram of number of words per sentence, and you'll find immediately that it's written by an AI.
lol, I copied the first 4 paragraphs into gptzero and it confirms, 100% ai generated. Which makes this submission against HN guidelines...
The only time I want to read something written by AI is when I am interacting with an LLM directly. I don’t know who these people are that think others want to read their slop. Even if the content is good, I can’t stand reading it.
It's written like a LinkedIn post, with the TED Talk pauses for effect. I've come to filter out anything written in that style.
It's written by an AI. That's why it's a bit shit.
I'd put my money on "edited by an AI". Some of it feels info-dense enough to be a person.
> They understand the business processes they’re digitizing.
I feel this has more importance than they think. Outside consultants would not have had this domain knowledge and would have spent months learning it. And then would have had to fix their mistakes because they misunderstood something (billed to the province, naturally)
We're having the users who understand the business processes use Claude to create clickable demos. Then these are presented to the engineering team to re-build. The users are loving it because they can ask for exactly what they want.
USDS and 18F proved that strong, internal technical leadership and dedicated teams make this work not AI tools.
The shortlist came down to four major consulting firms.
This is the problem in a nutshell. Those firms are structured to extract money from their customers, not to produce useful work. The fact that anyone is signing contracts with them any more blows my mind.
I wonder how much of the needs satisfied by this software in Alberta is also needed by other states? Yes, it's great that they saved money and built what they need so quickly, but a catch with govt is that they build out these proprietary processes that need more bespoke software.
AI seems somewhat orthogonal. It may have sped up the timeline, but the people using it need to be competent in the first place.
Well, an AI will be using it of course. The people will be using the AI.
Alberta needs to leave Canada. Imagine having the profit from all of your natural resources leave the state to fund social programs that don't benefit you at all. The rest of Canada does nothing for Alberta.
I think the use of AI is really missing the point here. The point is that small in-house teams can deliver a lot more quickly and to a higher quality and at a lower cost than large outsourced teams from the big consultancy companies. I've seen this over and over again (the problem is that large organisations often prefer to go the slow and expensive route with the big consultancy companies for a complex variety of reasons). So it would be like an article saying "our small inhouse team using VS Code did a much better job than a big outsourced consultancy using MS Visual Studio - isn't VS Code awesome".
I'm using AI every day (and it's not really about AI), but:
Anyone else closed the article immediately after seeing the low-taste, sloppy image at the top?
How do you call this aesthetic? "Futuristic vomit"? AKA "Generate image of: code blocks, AI-brain image, diagram, smiling guy and bunch of other crap. Make it look cool and futuristic, make no mistakes"?
I searched this report for "test". Result: none found.
> what if a small team of public servants, equipped with modern AI development tools, built the replacement systems themselves?
Next: bridges and brain surgery.
On the one hand, yes.
On the other, procurement is so broken, that if their inhouse team is only marginally better, it's a win.
> procurement is so broken
Anecdata: while i was in a tiny tiny software company, we got an in at a large auto manufacturer. They said they had been trying to get someone to do that job for like 2 years.
The job was of the 'two people 3 months' magnitude. The procurement system was also of the 'two people 3 months' magnitude so we simply gave up.
In the article's case, they could have done this even before coding assistants. It would have cost the estimated 5 million instead of 850k, but that's still 10x less than the 54 million.
Yes, I think the innovation here is social as much as technical. The AI Magic Sprinkle fairy dust on the proposal to solve internally, is what grants permission to go ahead. It both increases the chances of actual success and gives something external to put the blame on if something fails.
I totally agree.
As flawed as this new approach might turn out to be, the traditional approach may (or may not) have an even worse probability of success.
Mm, did you read the part where they have hundreds of people in and reporting bugs and feature requests every week? Testing is probably too granular a concept for the target audience of this blog post.
This is really awesome work. Solving these legacy IT problems in government is under appreciated. Love that you all were able to accomplish this.
Super disappointed to see most of the comments just complaining about AI and not engaging with the contents of the article.
When the article is written by AI too, it makes reading it an exercise in trying to distil it back to what you think it was prompted with.
I don't see any reason to engage with obvious slop, and I wish people didn't submit it to HN.
Thankfully they added that horrible AIslop image right in the beginning of the article, so I knew instantly to not trust the author on anything.
Well, Nate Glubish seems to be the Minister of Technology and Innovation of Alberta.
Bet Nate Glubish didn't write a single word. Not even the prompt for the "AI".
It was some intern...
The only thing this piece of information does to me is lower my respect for Alberta.