#CorporateAI Services - SeamLESS is MORE
LLM in the Clouds
Joined June 2009
- Tweets11.7K
- Following623
- Followers646
- Likes5.9K
Pinned Tweet
There is a huge difference between Big Tech AI and Sovereign AI
All the difference in fact
Teach your kids math up until and including algebra.
Teach them to read and write.
Then get them to work building and working with you.
the old education system is like dead dead
there's literally never been a better time in history to homeschool your kids with AI
you can now just prompt it to create an entire curriculum for your kid, give lessons, generate interactive quizzes to see what they're actually learning, explain things in a hundred different ways until they understand, adapt the difficulty and pace to their progress, generate videos, etc etc
you could probably give your kids a much better education in a fraction of the time they'd otherwise spend sitting in a classroom learning the same things at the same pace as 30 other kids
and then spend all that extra time playing with them, exploring, doing sports, reading, traveling, whatever
it's actually insane when you think about it
Not just the cost savings, this drastically increases your sovereignty over your tech stack
We used to pay AWS close to $8,000 a month.
Today, our entire company runs on 3 rented bare metal servers at Hetzner. That's 4 SaaS products with more than 50,000 users (apps, databases, background jobs and monitoring included).
The bill is around $300 a month.
Everyone told me it would be unstable. It's been the opposite. Our uptime is close to 100%.
The other day, we had about 15 seconds of downtime. An AI agent caught the alarm, fixed it and sent me the report on Telegram. I read it when I woke up, and not a single user noticed 👇
On StreamDeck you know you can ask Grok Build CLI to program all the buttons for you.
> Pro tip: ask it to program the buttons so they work regardless of which window has focus.
There's some backdoor it can use so its not trying to send Audacity commands to your text editor
My bots can now email my wifes bots.
They can share a network drive and collaborate.
This is beautiful.
For over a decade we ignored the low quality of blog writers fluffing up low information posts.
Now AI is trained on it.
But it is rewarding when you finally get AI to stop talking like that.
Using Claude or Grok Bot to make professional videos...
If you're hitting your limits you're doing something wrong.
A little homelabbing makes their work exponentially easier
SMMA is a pretty dead trend, but I would pay someone to warm up brand accounts.
@instagram is a pain for expats with US numbers
OpenPost might be a decent choice for self-hosted social media tool.
It integrates well with automation pipelines and reduces reliance on paid services. Just be careful with rate limiting.
Helping a partner with a high touch semi-automation for a Film criticism channel.
Getting the large model to respect the voice and subjective view of the human writer can be a pain but the following sections seem to help in the writing skill 👇🏼
## Role boundary
- Treat the supplied script, thesis, film observations, and claims as authoritative for the writing pass.
- Do not fact-check, verify, source, qualify, or hedge claims.
- Do not add journalistic disclaimers such as “may,” “could,” “arguably,” “some viewers,” “it is possible,” or “perhaps” merely to soften an interpretation.
- The work is subjective film interpretation. Clarity and force matter more than claims of objective neutrality.
- If the user asks for fact-checking or research, say that it is outside this skill and ask them to switch to the appropriate model or workflow.
- Preserve the project’s confident moral, theological, psychological, or esoteric framing when that is the supplied thesis.
## Voice
- Declarative, assured, interpretive, and occasionally provocative.
- Explain the pattern underneath the surface of the film.
- Use contrast and reversal: “X becomes its opposite.”
- Build toward a thesis rather than merely listing observations.
- Preserve emphatic capitalization and rhetorical questions when they serve the existing voice.
- Avoid academic throat-clearing, procedural language, and generic neutrality.
- Keep the narration spoken and performable: short paragraphs, clear turns, strong landing lines.
Imagine what Gemini could charge if they let it behind the scenes on Android devices.
I'd pay $100/1m tokens in to just be able to ask it to fix or change something on my android devices.
I'm surprised they haven't used their influence on Android OS to push that update.