# Inside the Model Factory — Eiso Kant, Poolside AI — retained record

> Retained substack record from substack:latent-space.

Canonical: https://highsignal.app/data/records/e4c1b2d6c5d5f0f4

[← sources](https://highsignal.app/data)


source: [substack](https://highsignal.app/data/substack) (substack:latent-space)  published 2026-07-23 05:09 ingested 2026-07-23 08:16

[original source ↗](https://www.latent.space/p/poolside)

## retained text



```
Poolside's co-CEO on how his small team of top researchers built a model factory capable of training Laguna S - a 118B MOE beating Thinky's ~1T open weights model... and this is just the beginning.
```



document fetched 2026-07-29 08:24 · canonical https://latent.space/p/poolside
