Dear {occupant/resident},
Our state-of-the-art advanced systems have determined that you have committed/are about to commit a crime of {fill-slot}.
Authorities and/or your nearest neighbors and/or family will be notified (unless you are a member of our Elite Membership level).
This is a notification required by local ordinance {cite}. Please reply to this message to agree and/or confirm receipt.
Regards.
{insert pricing-link + call-to-action}
I pay for Cursor, OpenAI, and Anthropic. I was paying for Google Gemini as well, but it was returning too many errors so I canceled it. I also pay for Google office, Microsoft office, and Adobe subscriptions. They inject their own AI into their services, but I end up ignoring them or turning them off.
Mostly use it for coding in Cursor, but occasionally for research into the state of AI and to make MCP extensions. It’s been worth the investment so far, given how much more of the mundane coding tasks get done by supervising it. I also had it update a Wordpress theme because I had no interest in learning the innards.
I never let them loose in ‘agentic’ mode, as they inevitably destroy all the work. I can run decent-sized models locally through lmstudio and Cline, but they’re much slower than just using Cursor and a cloud model.
Outside coding, the only usable one I’ve found is Adobe Firefly, accessed inside Photoshop (to remove material) and Illustrator (to generate simple SVGs and icons from prompts).
Every single other one, when I’ve put it to a non-coding use has been a pile of slop. If all LLMs go away tomorrow, the only one I’ll miss is the Adobe SVG creator.
Plot: a rival publisher hires a killer to murder a successful author over the copyright.
The dessicated forms of past knowledge-seekers had turned into mortar, holding up the precarious walls of ancient books.
“Join us,” they whispered.
“Eww,” I replied. “Gross.”
I had my photographs.
Slow news day, or are the Summer interns heading back to school?
Hey, down in the front! The rest of us paid good money for the magic clown show.
Until the can of food is opened, and then you are only a distant memory.
Overused HDR and AI have led some of us to question the veracity of any of this.
The painter’s hands were seriously scarred setting up the model.
Both Google and Microsoft tried with actual underwater data centers. It looked like it was feasible, but repairing something when it’s deep underwater was not feasible.
Near water, there will be environmental issues – similar problems to factories and nuclear power plants, debris in intakes and warm water affecting marine life.
Basically, the solution is to build more efficient or less compute-intensive software, but that’s not where we all seem to be heading right now.
Noooooo… how else can we rebuild the garage door opener firmware from 1992?
BRB. Building a wearable that sends out an ultrasonic screech to blow out nearby microphones.
Going to call it “Blow Me.”
Edit: looks like they already exist: https://www.newscientist.com/article/2328652-voice-jammer-stops-anyone-from-recording-you-speak/
Also, paper with more detail: https://arxiv.org/pdf/1904.08490v1
If you want to talk bigger picture… they built a pedestrian/bike lane with zero access/amenities at either end (unlike the Bay Bridge). No staging area to load/unload your bike, no parking, no bathrooms, no water fountains. Good luck finding all-day parking on city streets in Pt. Richmond or… San Quentin.
Here are the directions: https://marinbike.org/news/getting-to-from-the-r-sr-bridge-pathway/
Once approaching/getting off the bridge, if commuting by bike, there is no direct connection to the Bay Trail. So anyone living in Marin and wanting to commute to, say, Berkeley or El Cerrito so they can get on BART (or even the Richmond Ferry) has to risk going through heavily industrial areas or dicey parts of Richmond. On the Marin side to/from Larkspur Landing, you had to ride unprotected on the shoulder of the freeway!
Only way to use it for commuters on either end would be to park and ride, but again, no parking and ride facilities. And there’s any wonder more people don’t use it? There’s wide open space at either end of the bridge to build staging areas, especially on the Richmond side right near the Toll Plaza, but nobody wants to make it easy. It’s such a gauntlet I’m amazed that many people use it.
The solution to traffic congestion is to make public transit and alternate forms of transportation more cost-effective, functional, and convenient. That includes offering easy transition/transfer points. It isn’t to open more driving lanes. They’ve known this since the days of Robert Moses in NYC, but keep doing it.
A few suggestions:
Some of those components may end up costing a lot to operate. You said you’re doing it as a portfolio piece. May want to create a spreadsheet with all the services, then run a cost simulation. You can use the AWS Cost calculator, but it won’t be as flexible doing ‘what if’ scenarios. Any prospective employer will appreciate that you’ve given some thought to runtime pricing.
You may want to bifurcate static media out and put them in S3 buckets, plus put a CloudFront CDN in front for regional scaling (and cost). Static media coming out of local server uses up processing power, bandwidth, storage, and memory. S3/CloudFront is designed for just this and is a lot cheaper. All fonts, js scripts, images, CSS stylesheets, videos, etc. can be moved out.
Definitely expire your CloudWatch log records (maybe no more than a week), otherwise they’ll pile up and end up costing a lot.
Consider where backups and logs may go. Backups should also account for Disaster Recovery (DR). Is the purpose of multiple AZs for scaling or DR? If for DR, you should think about different recovery strategies and how much down-time is acceptable.
Using Pulumi is good if the goal is to go multi-cloud. But if you’ve hardcoded Aurora or ALBs into the stack, you’re stuck with AWS. If that’s the case, maybe consider going with AWS CDK in a language you like. It would get you farther and let you do more native DevOps.
Consider how updates and revisions might work, especially once rolled out. What scripts will you need to run to upgrade the NextCloud stack. What are the implications if only one AZ is updated, but not the other. Etc.
If this is meant for business or multiple users, consider where user accounts would go? What about OAuth or 2FA? If it’s a business, they may already have an Identity Provider (IDP) and now you need to tie into it.
If tire-kicking, may want to also script switching to plain old RDS/Postgres so you can stay under the free tier.
To make this all reusable, you want to take whatever is generated (i.e. Aurora endpoints, and save everything to a JSON or .env file. This way, the whole thing can be zapped and re-created and should work without having to manually do much in the console or CLI.
Any step that uses the console or CLI adds friction and risk. Either automate them, or document the crap out of them as a favor to your future self.
All secrets could go in .env files (which should be in .gitignore). Aurora/RDS Database passwords could also be auto-generated and kept in SecretsManager and periodically rotated. Hardcoded DB passwords are a risk.
Think about putting WAF in front of everything with web access to prevent DDOS attacks.
This is a great, learning exercise. Hope you don’t find these suggestions overwhelming. They only apply if you want to show it off for future employers. If it’s just for personal use, ignore all the rest I said and just think about operating costs. See if you can find an AWS sales or support person and get some freebie credits.
Best of luck!
If you have control of the server or platform serving the content, could look into “robots.txt” and “tarpits.” There are a few, but one example is Nepenthes: https://zadzmo.org/code/nepenthes/
If you just own the domain and it’s hosted elsewhere, you could set it up to go through CloudFlare DNS. They have a one-button scrape-stopper: https://blog.cloudflare.com/declaring-your-aindependence-block-ai-bots-scrapers-and-crawlers-with-a-single-click/