---
title: "doubleAI asked what would happen if a code agent were inserted into the training loop that could… — doubleAI — DecisionGateway"
description: "investigation by doubleAI, recorded recently — verbatim quote and source citation."
canonical_url: "https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6"
meta:
  "og:description": "investigation by doubleAI, recorded recently — verbatim quote and source citation."
  "og:image": "https://dgw.tmarket.io/og.jpg"
  "og:image:alt": "White network of connected nodes crossed by a timeline on black"
  "og:image:height": "1024"
  "og:image:width": "1024"
  "og:title": "doubleAI asked what would happen if a code agent were inserted into the training loop that could… — doubleAI — DecisionGateway"
  "twitter:image": "https://dgw.tmarket.io/og.jpg"
---

> Index of every page on this site: [llms.txt](https://dgw.tmarket.io/llms.txt)

investigation

# **doubleAI asked what would happen if a code agent were inserted into the training loop that could change the data, the objective function, the weight update and even the architecture itself during learning.**

A investigation by [**doubleAI**](https://dgw.tmarket.io/actors/01a1e1f3-e4af-701c-8fdc-eddc768cc18d). Statements are as reported by the cited publishers — not verified by DecisionGateway.

> *ב-doubleAI שאלו מה יקרה אם נכניס ללולאת האימון סוכן קוד שיכול לשנות תוך כדי למידה את הדאטה, את פונקציית המטרה, את עדכון המשקולות ואפילו את הארכיטקטורה עצמה.*

mathyaiwithmikePublished 2026-11-08 [ Read source](https://t.me/mathyaiwithmike/1306)

[**doubleAI dossier **](https://dgw.tmarket.io/actors/01a1e1f3-e4af-701c-8fdc-eddc768cc18d)

## **Discussion**

## **Related claims**

[**The agent implants "code layers" into the network — small programs that read tokens or internal representations, and whose output re-enters the computation alongside attention and MLP.**post author · statement](https://dgw.tmarket.io/claims/01a1e1f6-8f42-7b81-9ede-400073e8096b) [**We are entering an era of self-improving code: coding agents will write the programs, and CEMs will ensure the machine runs them optimally.**post author · prediction](https://dgw.tmarket.io/claims/01a1e1f3-e528-7db4-8ba4-01fbbc3aaa79) [**Maxwell, a tool by doubleAI, does not write code with you; it lives inside the program at runtime, between the code and the silicon, examining flowing data and rewriting algorithms and machine instructions 'on the fly' to maximize performance (Superoptimisation).**post author · statement](https://dgw.tmarket.io/claims/01a1e1f3-e4de-702c-b62d-942f81744141) [**On algorithmic tasks (CLRS-Text), ten such layers raised an 86M-parameter model from 55% to 84%, after regular training got stuck at 55% and even training twice as long did not move it.**post author · statement](https://dgw.tmarket.io/claims/01a1e1f6-8f8e-7a10-a02f-2baf8c6d789e) [**To know where to implant, the agent computes the loss gradient for each token according to its representation in each layer.**post author · statement](https://dgw.tmarket.io/claims/01a1e1f6-8f4f-7234-a827-16c47834e007) [**The researchers show that training leads the model to learn an efficient solution path, even though it receives no extra reward for choosing that path.**החוקרים · investigation](https://dgw.tmarket.io/claims/01a1e1f1-8d4e-7407-9371-5176e7fb9053)

## **More claims by doubleAI**

No other claims by this actor on record.

[All claims](https://dgw.tmarket.io/claims)

```json
{"@context":"https://schema.org","@graph":[{"@id":"https://dgw.tmarket.io/#website","@type":"WebSite","description":"Decisions, events and falsifiable forecasts from 4,500+ Telegram channels — every claim cited. Read free.","name":"DecisionGateway","url":"https://dgw.tmarket.io/","publisher":{"@id":"https://dgw.tmarket.io/#identity"}},{"@id":"https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6#webpage","@type":"WebPage","description":"investigation by doubleAI, recorded recently — verbatim quote and source citation.","name":"doubleAI asked what would happen if a code agent were inserted into the training loop that could… — doubleAI — DecisionGateway","url":"https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6","breadcrumb":{"@id":"https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6#breadcrumb"},"isPartOf":{"@id":"https://dgw.tmarket.io/#website"},"potentialAction":[{"@type":"ReadAction","target":["https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6"]}]},{"@id":"https://dgw.tmarket.io/#identity","@type":"Organization","name":"DecisionGateway","url":"https://dgw.tmarket.io"},{"@id":"https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6#breadcrumb","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","name":"Home","item":"https://dgw.tmarket.io","position":1},{"@type":"ListItem","name":"Claims","item":"https://dgw.tmarket.io/claims","position":2},{"@type":"ListItem","name":"doubleAI asked what would happen if a code agent were…","position":3}]},{"@id":"https://dgw.tmarket.io/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6#/schema/quotation/1","@type":"Quotation","dateCreated":"2026-11-08","isBasedOn":"https://t.me/mathyaiwithmike/1306","text":"ב-doubleAI שאלו מה יקרה אם נכניס ללולאת האימון סוכן קוד שיכול לשנות תוך כדי למידה את הדאטה, את פונקציית המטרה, את עדכון המשקולות ואפילו את הארכיטקטורה עצמה.","url":"/claims/01a1e1f6-8f36-7022-92ef-e27c0d8d64f6","spokenByCharacter":{"@type":"Person","name":"doubleAI","@id":"/actors/01a1e1f3-e4af-701c-8fdc-eddc768cc18d"}}]}
```

## Sitemap

See the full [sitemap](https://dgw.tmarket.io/sitemap.md) for all pages.
