Johannes Hagemann ✓

@johannes_hage · sf · joined 23 Mar 2017

co-founder/cto @PrimeIntellect | open superintelligence infra, longevity, techno-optimism

10 817Followers

2 764Following

1 776Posts total

974KViews on collected posts

Derniers posts

Johannes Hagemann ✓@johannes_hage26 Aug

https://t.co/HnB44vwm4s 5K views · 71 likes · 3 reposts · 2 replies Open on X →

Johannes Hagemann ✓@johannes_hage17 Aug

megakernel implies the existence of at least: gigakernel terakernel petakernel exakernel ultrakernel, and kilokernel who's building this? 32.9K views · 191 likes · 7 reposts · 15 replies Open on X →

Johannes Hagemann ✓@johannes_hage15 Aug

which way anon? https://t.co/LylTTelSEH 8.8K views · 102 likes · 5 reposts · 5 replies Open on X →

Johannes Hagemann ✓@johannes_hage15 Aug

We ran the largest open experiment on how frontier models do AI research. 100+ autonomous runs across 10+ models, sandboxed on 8xH200s for up to 8 days, iterating on the nanoGPT optimizer track. Best runs closed 82% of the gap to a record built by dozens of humans over months. 590.4K views · 1.9K likes · 213 reposts · 74 replies Open on X →

Johannes Hagemann ✓@johannes_hage7 Aug

multi-agent support is live in prime-rl + verifiers you can now train and eval setups like: - Agentic Judging — solver traces are graded by a judge - Self-Play — a model playing against itself - User-Sim — a user agent interacting with an assistant https://t.co/Zvy44y9WJP 21.1K views · 242 likes · 14 reposts · 7 replies Open on X →

Johannes Hagemann ✓@johannes_hage7 Aug

Today, we’re extending our RL stack beyond individual agents to multi-agent systems. You can now express arbitrary agent interactions and train them. https://t.co/cKBHB430f4 228.5K views · 935 likes · 78 reposts · 48 replies Open on X →

Johannes Hagemann ✓@johannes_hage5 Aug

Prime Agent combines several research ideas we've been exploring over the past few months: - Recursive Language Models-native programmatic tool calling - Persistent multi-agent orchestration ⁃ A self-improving Continual Harness Try it out: https://t.co/aeGfTdkdNh https://t.co/c 66.8K views · 810 likes · 47 reposts · 23 replies Open on X →

Johannes Hagemann ✓@johannes_hage8 Jul

We've raised $130M to build the Open Superintelligence Stack and make frontier AI infrastructure, once locked behind the walls of the big labs, accessible to every company. https://t.co/oU1yDtZwKe 20.5K views · 233 likes · 13 reposts · 33 replies Open on X →

Face aux comptes de taille comparable

8 posts des 90 derniers jours, à côté de la tranche de 10K–100K abonnés. diffusé à plus de monde que les comptes de taille comparable.

Vues médianes27 019ce compte1 098médiane pour 10K–100K

Portée, %249.78%ce compte3.91%médiane pour 10K–100K

Engagement, %1.26%ce compte1.66%médiane pour 10K–100K

Indicateur	Ce compte	Médiane pour 10K–100K	Rapport
Vues médianes par post	27 019	1 098	24.6×
Portée (vues ÷ abonnés)	2.5× audience	3.91%	63.9×
Taux d'engagement	1.26%	1.66%	0.76×

Autres comptes de cette tranche → Comparer avec un autre compte → Comment ces repères sont établis →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

20.5K8 Jul

66.8K5 Aug

228.5K7 Aug

21.1K

590.4K15 Aug

8.8K

32.9K17 Aug

5K26 Aug

Last 8 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

1.37%8 Jul

1.32%5 Aug

0.48%7 Aug

1.25%

0.39%15 Aug

1.27%

0.66%17 Aug

1.53%26 Aug

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.66%.

What the audience does

Likes63.4%4 466 in total

Reposts5.4%380 in total

Replies2.9%207 in total

Quotes2.1%145 in total

Bookmarks26.2%1 848 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.