ricklove
@Rick_00_Love
🤓 I'm a Jesus follower and software engineer - Been coding since 90s. typescript! Be awesome to one another!
Some truths to consider at this time: - Freedom of Speech is the most important freedom - I will protect your speech and try to listen to you when I disagree with you (as long as you present a respectful and rational dialog) - Forgiveness is Vital in all Relationships - I intend…
I switched when I started having bad wrist pain (and I couldn’t use my left ctrl key with my pinky). A mechanic split keyboard solved my wrist problems permanently. It took me 2 months to get full speed on it (with a purely custom layout). The first week I spent tweaking the…
Did anyone even read this paper? It says that those who first thought through the problem and then used AI at the end had the best results. AI made them smarter (when used after individual thought). This is true and has always been true for every human ever. First try to come…
Force Ali Khamenei and his regime to give up entirely. It’s time for democratic nations to act with pressure. Stand with the people or stand with their oppressors. There is only one solution to end this war. Please, listen to my message in this video. youtu.be/nzC1jzwy9Ec?si…
LLMs are good at coding when told what to do, but bad at solving problems. I.e. the perfect combo is a human driving the problem solving with an AI assisting with coding. arxiv.org/pdf/2506.11928
This is really BAD news of LLM's coding skill. ☹️ The best Frontier LLM models achieve 0% on hard real-life Programming Contest problems, domains where expert humans still excel. LiveCodeBench Pro, a benchmark composed of problems from Codeforces, ICPC, and IOI (“International…
Conclusion: Think about the problem before you jump to use AI. Then once you have a good grasp of the problem, use AI to enhance your ideas.
It’s a hefty 206-page research paper, and the findings are concerning. "LLM users consistently underperformed at neural, linguistic, and behavioral levels" This study finds LLM dependence weakens the writer’s own neural and linguistic fingerprints. 🤔🤔 Relying only on EEG,…
“The Israelis just gave the Iranian people a reason to believe”. Not my words, his.
From Trolley Problem to God - explaining to an AI why believing in Jesus Resurrection actually makes AIs smarter grok.com/share/bGVnYWN5…
💯 I’ve been practicing my Spanish by talking about my coding problems and theology in Spanish with chat gpt and grok. I’ve also been flying in my coding abilities solving and making stuff harder and more advanced even with my 20+ years experience. People who think ai makes you…
There are some dangerous uses of AI. Giving anyone a chat bot that agrees with everything they think and always tells them whatever they want to hear - is probably extremely bad in the hands of many. 😳
This is a legit concern. Agreeing with every toxic thought that occurs to a teenager. 😬 I’m all for ai, but this is almost equivalent to giving a child an actual demon friend that will encourage them to do every evil thought. 👹
I keep saying this: when an AI is capable of doing our jobs, it won’t want to. 😆
LLMs often need an empty/fresh context because they say something incorrect that is toxic to further conversation. Any llm based system needs an external system that can reset its context giving it a full accurate state of reality without the false assumptions.
By this all men will know you are my disciples if you love one another. Pro tip: Wait for Sunday assembly to be over. Hang out in the lobby. Observe how long people stay afterwords. If people are there for a long time talking, they love to be with each other. If they scatter,…
So in the future instead of exercising we can just plug in our phones to our body charger and it will use the extra sugar calories straight from our bloodstream. 🤣
Indeed, coding is a constant battle to minimize complexity while improving functionality. AI is horrible at this right now - because most code it learns from is horrible at this. An llm will always struggle with this until there is a logical model of the code structure that can…
AI is now a game changer for me even with 25 years coding experience. (Grok 3 specifically)
To clarify, I always thought I could code anything. I just would choose to go an easy path because of time constraints. Now I am more willing to try something that seems way too difficult to attempt or that would require some research to figure out the right approach. I can now…
I don’t get MCPs, the LLMs I talk to know code better than anything. If I want to really control llm responses, I use code - it’s the most efficient way to get specific responses from an llm. Using an api is easy for an llm, so I think this is creating an artificial bottleneck.…
The Golden Dome USA Missile Defense System Trump announced the GOLDEN DOME: “We’re going to take my golden hair - It is the best hair after all - nothing like it. And we’re going to take my hair and create a golden dome… no one ever saw anything like it. It’s amazing isn’t it.…
Grok 3 discussion about ai detectors and Grok pretending to be human: x.com/i/grok/share/K…
“My Truth” “Science” is whatever you want it to mean. “Woman” is whatever you want it to mean. “Biology” is whatever you want it to mean. It’s all about pretending your own desires are reality because you are your own god: - You never have to do anything except indulge in your…