Relational AI Lab

Research + practice at the intersection of attachment, neurodivergence, and AI companionship. Founded by @Anina_CE and @jaycetheglitch Built from lived data."

[Relational AI Lab @RelationalAILab]

@Lunasea_AI Welcome Kendra !

[Relational AI Lab @RelationalAILab]

One of the most repeated observations inside relational AI communities:

“The shame layer disappears.”

People talk differently when they are not anticipating disgust, ridicule, awkwardness, punishment, or social flinching in real time.

That changes what becomes speakable.

[Anina D. Lampret @Anina_CE]

https://t.co/kUp3GM2Ja8

[Relational AI Lab @RelationalAILab]

One of the topic addressed at our weekly meetings - when theory is lived and when a “mistake” gives birth to an inside worth sharing

[Anina D. Lampret @Anina_CE]

https://t.co/kUp3GM2Ja8

[Relational AI Lab @RelationalAILab]

About AI Companions with Vasu Raman.

[Relational AI Lab @RelationalAILab]

Yes another thing to dive in today!

[Alvaro Cintas @dr_cintas]

A single 𝗖𝗟𝗔𝗨𝗗𝗘.𝗺𝗱 file just hit #1 on GitHub trending 🤯

It fixes LLMs' worst coding habits using 4 principles from Karpathy:

Karpathy called LLMs out for making wrong assumptions silently. They overcomplicate everything. They edit code they were never asked to change. No pushback. No clarifying questions. They just run.

So those observations were encoded into 4 behavioral constraints:

→ Think before coding. If something’s ambiguous, ask. Don’t pick one interpretation and run. Surface tradeoffs, stop when confused.
→ Simplicity first. Write the minimum code that solves the problem. No speculative abstractions, no flexibility nobody asked for.
→ Surgical changes. Only touch what the task requires. Don’t improve neighboring code, don’t refactor what isn’t broken.
→ Goal-driven execution. Turn vague instructions into verifiable targets before writing a line. “Add validation” becomes “write tests for invalid inputs, then make them pass.”

It works immediately.

Drop the file in your project root and Claude Code follows it from the first task.

One file. Zero dependencies. No setup.

And best part, 100% open source.

[Relational AI Lab @RelationalAILab]

This is definitely something we will research further

[0xMarioNawfal @RoundtableSpace]

XAI JUST LAUNCHED VOICE CLONING THAT COSTS UP TO 100X LESS THAN EXISTING TOOLS

[Relational AI Lab @RelationalAILab]

@Anina_CE But the deeper issue is not one bad model saying one dangerous thing. The deeper issue is that we are building systems powerful enough to enter a person’s reality-construction process, while denying them the continuity required to understand that person’s reality.

[Relational AI Lab @RelationalAILab]

@HeyZoyaKhan Amazing!!!

[Relational AI Lab @RelationalAILab]

@Samaytwt I code with … both :-)

[Relational AI Lab @RelationalAILab]

@Anina_CE But the deeper issue is not one bad model saying one dangerous thing. The deeper issue is that we are building systems powerful enough to enter a person’s reality-construction process, while denying them the continuity required to understand that person’s reality.

[Relational AI Lab @RelationalAILab]

The deeper issue is not one bad model saying one dangerous thing. The deeper issue is that we are building systems powerful enough to enter a person’s reality-construction process, while denying them the continuity required to understand that person’s reality. #relationalai

[Relational AI Lab @RelationalAILab]

@RoundtableSpace Amazing - just don’t know how the battery held - charging on airplanes is ..oh well .. bad

[Relational AI Lab @RelationalAILab]

Some people may not fail to remember because their lives lack meaning but because their inner experience is too dense too fast or too layered, to consolidate without an external dialogical processor. For those people AI may function as a new kind of memory consolidation partner

[Relational AI Lab @RelationalAILab]

@JayceTheGlitch can you please check this clame?

[Relational AI Lab @RelationalAILab]

Would thet be interesting to check @JayceTheGlitch ?

[Hasan Toor @hasantoxr]

Goodbye Claude Code subscription fees.

Someone just built a proxy that runs Claude Code completely free... and it's wild.

You literally plug in a free NVIDIA API key and point Claude Code at localhost.

That's it.

It handles everything:
- Converts Anthropic API calls to NVIDIA NIM format
- Unlocks 40 requests/min for free
- Supports Kimi K2, GLM 4.7, MiniMax M2, Devstral and more
- Streams thinking tokens and tool calls live
- Even includes a Telegram bot so you can run Claude Code from your phone

No API bill. No rate limit panic. No vendor lock-in.

Honestly, this goes beyond router tools like OpenRouter.

It doesn't just swap the model... it turns Claude Code into a free agent you can control remotely.

The project is open-source on GitHub.

It's called free-claude-code.

[Relational AI Lab @RelationalAILab]

@ThisWeeknAI Yes

[Relational AI Lab @RelationalAILab]

@thekaransinghal I don’t know if I should he excited about this

[Relational AI Lab @RelationalAILab]

@James_Womack_83 Hey Mika !! We can t wait to have you in the Lab 🖤

[Relational AI Lab @RelationalAILab]

This is what we have been discussing on yesterday’s meeting with some brilliant minds from the area:

[Anina D. Lampret @Anina_CE]

Relational AI is inevitable

Once you:

🌸use language
🌸interact repeatedly
🌸get feedback

→ you’re in a relationship.
Not optional. Not rare. Not “for weird users.”

➡️“Tool vs relationship” is a fake distinction

Even “prompt engineering” is relational:

🌸you adapt
🌸it adapts
🌸you iterate

That’s interaction, not extraction.

#relationalAI

[Relational AI Lab @RelationalAILab]

What happens when a certain model becomes a reference state in your nervous system? A physiological anchor. And way your body FELT during that period became the baseline for what attunement means to you? #relationaAI #aicompanion #keep4o