HNHacker News
TopNewBestAskShowJobs

thejash

466 karma · joined October 7, 2008

http://joshalbrecht.com/

joshalbrecht@gmail.com

submissionscomments

How Offload Works: Inside the Rust CLI that sped up our tests by 6x

imbue.com·12 pts·thejash·
0

A case study in testing with 100+ Claude agents in parallel

imbue.com·68 pts·thejash·
54

Usefully run 100s of Claudes in parallel with mngr

imbue.com·21 pts·thejash·
0

Software as Bonsai

substack.com·2 pts·thejash·
0

AI should be Free Software

substack.com·3 pts·thejash·
0

Why agents matter more than other AI

substack.com·1 pts·thejash·
1

What's needed for AI agents to work

substack.com·2 pts·thejash·
1

Decentralization or Dystopia: Pick 1

substack.com·1 pts·thejash·
0

Show HN: Sculptor – A UI for Claude Code

imbue.com·176 pts·thejash·
85

Sculptor: Catch and fix issues as you code

imbue.com·61 pts·thejash·
3

Carbs, a hyperparameter optimizer that scales small experiments to LLMs

imbue.com·4 pts·thejash·
1

Open-source, sanitized evaluation datasets for models that reason and code

imbue.com·3 pts·thejash·
1

Infrastructure setup and open-source scripts to train 70B model from bare metal

imbue.com·325 pts·thejash·
41

Scaling Laws for Every Hyperparameter via Cost-Aware HPO

generallyintelligent.com·3 pts·thejash·
0

Jim Fan, Nvidia: On foundation models for embodied agents and scaling data

generallyintelligent.com·2 pts·thejash·
1

Noam Brown, FAIR: On achieving human-level performance in Diplomacy, and more

generallyintelligent.com·1 pts·thejash·
1

Nicklas Hansen, UCSD: On long-horizon planning and much more

generallyintelligent.com·1 pts·thejash·
1

Jack Parker-Holder, DeepMind: On open-endedness, offline learning, and much more

generallyintelligent.com·1 pts·thejash·
1

Chelsea Finn, Stanford: On the biggest bottlenecks in robotics and RL

generallyintelligent.com·1 pts·thejash·
1

Hattie Zhou, Mila, on supermasks, iterative learning, and fortuitous forgetting

generallyintelligent.ai·3 pts·thejash·
1

Generative AI startups we want to fund

outsetcapital.com·1 pts·thejash·
0

Minqi Jiang, UCL, on environment and curriculum design for general RL agents

generallyintelligent.ai·2 pts·thejash·
1

Oleh Rybkin, UPenn, on exploration and planning with world models

generallyintelligent.ai·1 pts·thejash·
0

VICReg: Tutorial and Lightweight PyTorch Implementation blog post

generallyintelligent.ai·1 pts·thejash·
0

ZeroCater (YC W2011) surpasses 100M in total sales

blog.zerocater.com·3 pts·thejash·
0

Join our hacker house for new YC companies

founderflat.com·44 pts·thejash·
37

Apply to AlphaLab: $25K and 5mo free office space

alphalab.org·40 pts·thejash·
34