All People
Reproducing all of Jürgen Schmidhuber’s papers (1990-2025) using an AI coding assistant.
Cool project by Yaroslav! It even reproduced the “World Models” paper by me and Schmidhuber (2018) using a toy environment, with a full VAE + RNN world model implementation.
Project: github.com/cybertronai/...Excited to share Sakana AI’s new #ICML2026 paper in collaboration with NVIDIA: "Sparser, Faster, Lighter Transformer Language Models" arxiv.org/abs/2603.23198
This work introduces new open-source GPU kernels and data formats for faster inference and training of sparse transformer LLMs:
🧵 Thread 👇If GitHub were built in:
Japan 🇯🇵
China 🇨🇳
North Korea 🇰🇵
The EU 🇪🇺For the past few years, humans have been doing “prompt engineering” to coax the best performance out of different LLMs. In this work, we explored what happens if we train an AI to do that job instead.
Link to our #ICLR2026 paper: arxiv.org/abs/2512.04388
Thread:Scaling up massive LLMs continues to yield incredible results. But to truly unlock their full potential, the next frontier is test-time compute and dynamic orchestration.It's spectacular to have followed David Ha's (@hardmaru) incredible career arc —MD of Fixed Income at Goldman Sachs —restarted his careerDon't miss David Ha @hardmaru's keynote at @ALifeConf #ALIFE2021 on "World Models and Attention for Reinforcement Learning"!Don't miss David Ha @hardmaru's keynote at @ALifeConf #ALIFE2021 on 'World Models and Attention for Reinforcement Learning'!
D
Recent Activity5 posts · 3 x-posts
Recent Activity
grep SOURCE=
hardmaru
@hardmaru.bsky.social
AgentTooling
hardmaru
@hardmaru.bsky.social
LLMInfraDeployment
hardmaru
@hardmaru.bsky.social
Infra
hardmaru
@hardmaru.bsky.social
AgentLLMFine-tuning
hardmaru
@hardmaru.bsky.social
LLMAgent
David Ha
@hardmaru
Highlights: David Ha transitioned from a managing director at Goldman Sachs to a career in AI research.
Worth reading: Shows the diverse background and career pivot of David Ha, an influential AI researcher.
Agent
David Ha
@hardmaru
Highlights: David Ha is giving a keynote on world models and attention for reinforcement learning at ALIFE 2021.
Worth reading: Highlights David Ha's influential work on world models and attention mechanisms in AI.
AgentMulti-modal
David Ha
@hardmaru
Highlights: David Ha gave a keynote on world models and attention for reinforcement learning at ALIFE 2021.
Worth reading: Highlights Ha's work on world models, a key concept in AI.
Agent
5 posts · 3 x-posts · All time