An advanced technician with strong Catechesis. You'll learn about technology, a little Catechesis. Founder and developer of and for PetraOS.
United States of America
Joined April 2024
- Tweets7.3K
- Following4.8K
- Followers3.7K
- Likes9.4K
Pinned Tweet
We have lift off, but the GitHub isn't quite ready to go public just yet. I just have some more hardware combinations to test, and I'm ALWAYS taking suggestions as to what YOU would like to see in a desktop OS!
Cate-Tech retweeted
When an AI agent fails at accounting or ops, it is not a reasoning issue; it is a deterministic math issue. Exact math is not a matrix probability outcome.
Let the model make the decisions. Let code do the math. How MDCC hits 100% accuracy:
The scapular is a sacramental, not a magic charm. Per the Catechism (1670), sacramentals don't confer grace like sacraments do; they dispose us to cooperate with it. The Sabbatine Privilege is a pious tradition with conditions, not automatic.
Sacramentals aren't magic. Per the Catechism (1670) they prepare us to receive grace and cooperate with it. The Church says attributing magical efficacy to signs is superstition (2111). The Sabbatine Privilege is a pious tradition tied to conditions (chastity per one's state, prayer), not an automatic release.
vatican.va/content/catechism…
catholic.com/magazine/online…
catholic.com/tract/the-brown…
Here is another gentleman that has discovered the fruits of running his own AI, as opposed to relying on subscription services and the cloud. People should follow more pages like this to learn more. Because it's not what you've been told, it's not what you think, and it's way more useful and better for our future than anyone ever led on
Cate-Tech retweeted
This is f*cking insane. Anthropic engineer:
"You're not supposed to prompt Opus 5.5
You're supposed to build a system that prompts itself. At Anthropic, 90% of our engineers have already built these systems"
In 30 minutes, she walks through exactly how Anthropic builds agents that remember, fix their own mistakes and get sharper with every run.
Better than any $1000-$2000 agent course I've seen.
Watch it, then read the article below ↓
Readers added context they thought people might want to know
Lamis never says "You're not supposed to prompt Opus 5.5” in here talk at AI Dev Con.
The caption misrepresents the video knowing most won’t watch the video all the way through and jump to the articles.
youtube.com/watch?v=tTcxVv…
Cate-Tech retweeted
THIS IS AN ACTUAL F*CKING GOLDMINE
Anthropic just shared one of the best tricks for Claude Code.
The idea: run Sonnet, Haiku 5.5 and Opus 5.5 at once, each one doing only the job it's best at.
Better results and way fewer tokens, because every task lands on the right model.
You just launch Claude Code like this:
claude --model sonnet --advisor opus
here's the split:
→ Sonnet 5.5 runs the session, writes the code and applies the fixes
→ Haiku 5.5 swarms the repo in parallel: search, grep and diff audits
→ Opus 5.5 stays on call as advisor for the big calls and signs off on the diff
→ you stop burning Opus context on work that doesn't need it
Opus only steps in at three moments:
→ before a plan locks
→ when the same error shows up twice
→ before calling it done
Sonnet builds. Haiku swarms. Opus reviews.
JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split
the full tree
> Sonnet 5.5 runs the main session and coordinates the work
> search maps the repo on Haiku 5.5, read-only
> grep finds call sites on Haiku 5.5 in its own context
> audit checks the diff on Haiku 5.5 in a throwaway worktree
> Opus 5.5 on call as advisor, reviews the synthesized diff before sign-off
> JEV routes micro-forks in milliseconds
paste the tree and this prompt into Claude Code ↓
"Reconfigure my Claude Code setup around this tree:
1. Run the main session on Sonnet 5.5 with Opus 5.5 as advisor.
2. Create three Haiku 5.5 subagents in .claude/agents: search (maps the repo), grep (finds call sites), audit (checks the diff).
3. Set isolation: worktree on audit so its checks never touch my working branch.
4. Add one rule to CLAUDE.md: delegate every repo search and test run to the Haiku 5.5 subagents to keep the main context small.
5. Route fast deterministic forks (tool choice, path selection, retries) through JEV.
Show me every change as a diff first. No edits until I say go."
↳ code.claude.com/docs/en/sub-…
save it, it'll come in handy.
Cate-Tech retweeted
Running a Qwen3.6-35B-A3B-class model (35B total, ~3B active) at usable speed on an 8 GB laptop GPU was not practical a couple of years ago. Dense models of comparable capability needed far more VRAM, and early MoE offload paths were usually too slow to be interactive. The combination of sparse MoE architectures plus engines that keep only hot experts and attention on the GPU while leaving the rest in system RAM is what makes the laptop result viable now.
The broader point stands: local AI is becoming practical on ordinary gaming-laptop hardware in ways that were unrealistic recently. The specific run is one clear data point for that shift.
This is how Catholic and non-Catholic Christians should communicate more oftrn.
@TECHdigression @LILindaLee38
I agree. I can discuss virtually anything and remain civil and respectful. Name calling and slinging arrows does absolutely nothing productive. Agree to disagree or you're just spinning your wheels getting absolutely nowhere except into a deeper hole.
Linda, please by all means speak up and chime in at will.
Again, thank you extending your concern and friendship!
All the best,
Lyman
Oh boy, I may get some heat for this... The Bible is the What and how; the Catechism is the why and for what reason.
This is oddly more pointed towards my Catholic friends. We need to be better about studying our faith, and the why's. Especially before we larp as professional apologists, and perhaps accidentally smear the name of Christ, or his Church.
Sometimes I think I'm losing my mind, but then I remember the rest of the world has lost its. I don't know what to do anymore besides pray. Everything else has just seemed so hopeless and in ruins lately.
Catholics and Protestants need to stop fighting each other online, and start paying attention to who's indocrinating children and young adults with non-christian ideology, and what that ideology means to the future. Like Sharia Law
Cate-Tech retweeted
Amen to all of this. I think X apologetics is generally a screwed space for all Christians. We should all wonder if even that would be by design. I've come to the conclusion that it is. I've been reading and watching, lurking lately.
A Catholic sees nothing but aggressively horrible Protestants and Reformists... Yet a Protestant sees nothing but aggressive and angry Catholics.
We are indeed so busy fighting each other, that we're missing a huge threat to both Western culture and Christianity. We could literally be online, debating apologetics with Muslims or making the case for Christ with them.
Instead, we just stay flippant which each other. So long as you have faith in the Triune God, you're a Christian. That's it, and that's all. We can save the "Is Rome completely necessary or not" arguments for any other day. We can also save the "is Sola Scriptura even valid" debates until we make headway elsewhere.
You know what I have noticed? Muslims following teens accounts, adult and otherwise. Casually commenting on Christianity, quoting "Surah" this or that. Going from the Quran and Hadith to make the case for Islam.
While we do what? Fight with each other, all of the time.
To my Catholic brothers and sisters... Protestants ARE Christians too. Even our own dogma recognizes that. We're supposed to come from a place of humility and chairity... That's also a scripturally and dogmatically sound.
I digress...
