Ryan Peters (@ryanirl) header
Ryan Peters (@ryanirl) profile picture

reverse engineering intelligent (learning) systems. rl, evals, safety @emulated_ai

San Francisco, CA
Joined February 2019
Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)Photo by Ryan Peters (@ryanirl)
I’m hiring researchers at @emulated_ai for roles in evals, RL, safety, and ML infra. Must be in SF or open to relocating. Competitive pay. Happy to hire people transitioning into the field. I also know 3-4 other teams hiring right now in SF, NY, and remote. If you’re cracked, DM me and I can probably place you.
49
29
2
373
31,793
How would a Waymo solve the trolly problem?
3
4
813
Ryan Peters retweeted
Qwen4 is coming. Deeper knowledge. Richer world understanding. Stronger coding and agentic capabilities. Built for real-world tasks — from complex coding and knowledge work to front-end builds and professional software workflows. Qwen4-Max is now in preview on Alibaba’s Token Plan. You can also get early access to the new model on QwenWork, Qoder, Qwen APP, and QwenStudio. Put it to the test on a real problem. Show us what you build. 👀 Token Plan: qwencloud.com/pricing/token-…
3
19
1
452
4,392
Ryan Peters retweeted
Two years ago, I started a side project solo on my RTX 3090 Ti. This year, it became an independent SIGGRAPH paper. Here’s how I got there: teaching myself ML, doing research, writing, navigating peer review, and advice for anyone taking a similar path. alexandergoslin.me/writing-a…
Introducing InfiniteDiffusion, my independent paper accepted to #SIGGRAPH2026! I have one RTX 3090 Ti. No funding, advisors, or team. By day I'm a new grad SWE at Walmart. The paper has two main contributions: - InfiniteDiffusion: a new approach to infinite generation with diffusion models. - Terrain Diffusion: the world’s first learned procedural terrain generator. Here’s why this matters, and how they are connected. 🧵
25
74
2
957
35,373
Proud for @emulated_ai to be a science partner on this!
Today we’re releasing GDPx, a benchmark for measuring agentic capabilities across the most valuable domains. GDPx is a true frontier-level benchmark: even the best model scores just 12% (1/7)🧵
7
768
What is the smallest set of neurons you can adjust to reliably and robustly get Qwen3.8-27b to substitute your name in for everything? The smallest I have found for 'Ryan' is 12: L21N1097-0.10 L21N14604-0.15 L23N2759-0.10 L48N9077-0.15 L49N1412-0.10 L60N15241-0.15 L61N8815+0.05 L61N8987+0.50 L61N11372+0.10 L62N410-0.30 L62N13617+0.25 L63N10669+0.10
4
8
713
Down to 7 now 😅 L21N1097-0.10 L48N9077-0.25 L49N1412-0.20 L61N8987+0.45 L61N11372+0.15 L62N410-0.35 L62N13617+0.30
1
181
Grim reaper Qwen L39N12471+0.65
3
302
Happy Halloween everyone! L23N5199+0.75 gives you a very spooky Qwen3.8-27b.
3
9
499
Steering VMHvl Esr1+ neurons in male mice makes them aggressive and attack whatever’s in front of them, even an inflated glove. From Lin et al., 2011 nature.com/articles/nature09…
if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1 now you have evil qwen recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input
1
8
977
A small swarm versus a single agent in an autoresearch harness
45
34
15
997
125,028
New Kaggle theme 👀
3
455
Hey Claude, build a league of legends style game and then pit Astra against itself in a 5v5 using computer use.
3
1
9
750
It's actually quite cool to see them do the normal setup with both a support and adc bottom lane and such.
1
221
Build club penguin, make no mistakes.
1
9
368
Ryan Peters retweeted
total gnomic victory
Meet Mistral Large 4, aka Le Chonk. • 1T parameters, natively multimodal. 49B active. It is the best open weights model from US or Europe on aggregated benchmarks. • State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding. • Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure. • Available to all via API today. Working with cybersecurity partners privately. Open weights release end of October.
7
10
177
2,749