AI makes AI better.

A daily hub for work where AI is part of the loop that improves AI, each entry with its loop spelled out. Read more →

1/4

A loop in motion

Project · Google DeepMind

Gemini models propose code changes

From the labalphaBell Qwen Open27B post-training research, with capability tradeoffs measured.Explore the model → From the labModel ObservatoryMeasured Flint and Qwen traces with published decoder maps.Open the lab tool → From the labMiniMax Video BudgetVisualize sampled frames, two-frame groups and vision placeholders.Try the inspector →

Today's radar

27 Oct 2026Full radar →

8 picked from 125 candidates · 31 sources read · 72 loops indexed

01
Loop of the dayPaperarXiv26 Oct 2026Research automation
Infinity Machine: a self-improving AI agent for research
The loop: An AI agent modifies its own harness, prompts, and meta agent to solve and verify open scientific problems. These modifications pass through a kernel-enforced gate and are directly used by the agent to conduct further autonomous research.
Infinity Machine demonstrates a closed loop where an agent autonomously edits its own harness and meta-agent to achieve state-of-the-art results in quantum information.
loop fit 10/10Harish Santhanalakshmi Ganesanvia arXiv
02
PaperarXiv25 Oct 2026AI R&D evals
ParetoRSI-Gym: Can AI Agents Improve Language Models Across Multiple Domains?
The loop: An agent searches for and curates training data to post-train its own successor across multiple domains. The resulting model is evaluated on generalization benchmarks, and the results guide the agent in the next round of data selection.
loop fit 9/10Haojin Wang, Yangzhen Wu et al.via arXiv
03
Blog postNVIDIA26 Oct 2026Hardware and chips
How NVIDIA Minerva Turns Telemetry Chaos Into Peak GPU Yield
The loop: AI models analyze telemetry to optimize GPU yield and cluster resiliency in AI factories. This optimization improves the hardware infrastructure that trains and runs those same AI models.
loop fit 7/10Elizabeth Goodmanvia NVIDIA Technical Blog
04
PaperarXiv22 Oct 2026Tools and harness
Reproducible Evaluation as the Substrate for Safe Prompt Compression in Production Agents
The loop: An offline evaluation harness drives GEPA-based prompt optimization for a production conversational agent. The resulting compressed prompt is deployed back to the agent, shrinking its prompt surface while maintaining or improving task correctness.
loop fit 9/10Prabal Vashisht, Zheng Hu et al.via arXiv
05
Blog postOpenAI26 Oct 2026Tools and harness
How Ultrafast cut incident investigation from hours to minutes at OpenAI
The loop: Codex Ultrafast helps engineers investigate incidents and refactor code. This accelerates the development and security analysis of the underlying AI systems, including Codex itself.
loop fit 6/10via OpenAI
06
PaperarXiv23 Oct 2026Training data
On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models
The loop: A foundation model synthesizes its own curriculum of queries and privileged reasoning traces guided only by high-level topic descriptions. These synthetic traces are used for on-policy self-distillation, improving the model's performance across multiple domains.
loop fit 9/10Leon Sick, Lukas Hoyer et al.via arXiv
07
PaperarXivTools and harness
Harness Evolution as Learning: Approximation, Generalization, and Optimization Limits of Self-Improving Personal Agents
The loop: Personal agents evolve their own harnesses to optimize performance on tasks. The evolved harness is then deployed as the runtime environment for the agents, improving their execution of future tasks.
loop fit 8/10via Awesome-AI4AI
08
PaperarXivResearch automation
Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy
The loop: An autonomous system automates large-scale omnidisciplinary research tasks using a prompt economy. This automation improves the efficiency of the research process, which is then used to develop better AI systems.
loop fit 7/10via Awesome-AI4AI