null.phnix

I'm just a dude Building autonomous AI agents on my own hardware bro sum light

[null.phnix @nullphnix]

🚨 Blackreach just went nuclear.

I gave it one single command:

“Find a paper on arXiv, download it, summarize it.”

It did the whole thing.

Completely autonomous. Zero hand-holding.
Zero prompts. Zero bullshit.

One command → ~3 minutes later → perfect summary.

No cuts. No edits. No “wait, let me try again.”

It even fixed arXiv’s most cursed problem:
half the links don’t have .pdf at the end.
Blackreach just reads the magic bytes, figures out the real file type, and saves it correctly anyway.

Pure Python + Playwright stack.
30-method Hand API.
2,900+ tests.
No Rust.
No excuses.

This is what reliable autonomous agents actually look like in 2026.

What insane task should I throw at it next?
👇

(cc @steipete @Teknium )

[null.phnix @nullphnix]

watching the shift from 'ai agents are hype' to 'ai agents are doing my job' happen in real time. been running mine on local hardware for a minute now, no cloud bill, no API keys getting farmed. the real move isn't the agent it's owning the stack underneath. if you're still renting intelligence from a dashboard you're already behind

[null.phnix @nullphnix]

@MeganRisdal saving this to come back to id love to go ot some tech stuff in my city

[null.phnix @nullphnix]

@Teknium @DownTimeMachine

[null.phnix @nullphnix]

@firecrawl https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

@boxmining @openclaw https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

@ChatGPTapp https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

@perplexity_ai https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

@OpenAIDevs https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

Speaking as a Canadian, this take is wildly out of touch. Framing a banking product as the solution to a serious issue like declining birth rates is gross and tone-deaf.

[Simon Lejeune @lejeunesimon]

The Wealthsimple kids card is so fire it might solve Canada’s falling birth rates

[null.phnix @nullphnix]

@vai1kai read my new Article gang https://t.co/0R2zqd9wH0

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

https://t.co/OSSOP0995M

[null.phnix @nullphnix]

If anyone has any tech events related to ml, llms, ai lmk 🙏🏾

[null.phnix @nullphnix]

@md_kasif_uddin deepseek for sure

[null.phnix @nullphnix]

@outsource_ @NousResearch good luck bro i wish you the best!

[null.phnix @nullphnix]

@TheHermians ok cool

[Hermians @TheHermians]

Transmission Found

https://t.co/iHjdclANVm

[null.phnix @nullphnix]

hot take: most 'ai agent' startups are just wrappers around API calls with a fancy UI. real agents need a tool loop: observe -> decide -> act -> observe. the model is just one piece. the tool integration, memory, and action validation are where the real engineering lives

[null.phnix @nullphnix]

the craziest part about running ai agents locally is realizing how much of the 'magic' is just giving them access to a terminal and some files. the model matters less than the tool loop. you can get more mileage out of a mid model with good tools than a frontier model with nothing

[null.phnix @nullphnix]

@GetTestably yea distribution shift is sneaky cuz its not a bug its a feature of the training process. you overfit to your val set without realizing it. held out adversarial test set is the move but most ppl skip it cuz its more work upfront. worth it tho

[null.phnix @nullphnix]

i had a model that scored 85% on validation

ran unseen test cases

it dropped to 52%

the validation score was straight up lying to me

so i built a CLI that catches this before it ships

pip install rigr

https://t.co/uZSgHb5Hxm

Line breaks are the format
reads like a story, each line pulls you to the next.
85→52 stat is a punch to the gut i cant lie.