learning with yacine

Why DAGs Are Taking Over Auto-Research (w/ the Founders of Paradigma)

DAG + auto-research just work

May 25 • Yacine Mahdid

5

1

1:37:23

April 2026

Why Self-Distillation Is Taking Over LLM Post-Training (w/ the Researchers Behind It)

very pumped for this one shout out to Jonas Hübotter (ETH Zurich) and Idan Shenfeld (MIT) for blessing us with their knowledge

Apr 28 • Yacine Mahdid

1

1:31:19

The Science of Learning Math (and Anything Else) with Justin Skycak

you gonna have to take a whole day off for this one guys it's 3h wide with knowledge

Apr 15 • Yacine Mahdid

7

3:18:00

November 2025

within the chrysalis of prime intellect

Meditation at the prime intellect office on the shapes of artificial intelligence to come.

Nov 24, 2025 • Yacine Mahdid

7

1

how to stop having FOMO as a curious engineer: hold a thread

little worthless advice to learn passionately without boxing yourself

Nov 11, 2025 • Yacine Mahdid

17

September 2025

Hierarchical Reasoning Model assembly manual for toddlers

fun time for the whole family.

Sep 9, 2025 • Yacine Mahdid

6

2

1

August 2025

get your A in college and stop wasting your life

it's not complicated.

Aug 25, 2025 • Yacine Mahdid

5

2

how to stop feeling lost in tech: the wafflehouse method

you need to sit for this one for 48h

Aug 20, 2025 • Yacine Mahdid

116

14

13

muon optimizer explained to a toddler

there's no way you won't get it

Aug 19, 2025 • Yacine Mahdid

17

1

May 2025

Next Frontier for LLM is Quality Long Context

Which is much harder than you think.

May 26, 2025 • Yacine Mahdid

1

4

April 2025

How Minimax-01 Achieves 1M Token Context Length with Linear Attention (MIT)

I've dig into the internal of an MIT licensed MoE system that makes use of Linear Attention (Lightning Attention) to extend it's context length to 1M…

Apr 1, 2025 • Yacine Mahdid

1

March 2025

You shouldn't build fully autonomous agent.

That's a big no no.

Mar 10, 2025 • Yacine Mahdid

1