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 →
8 picked from 4 candidates · 30 sources read · 74 loops indexed
- 01Loop of the day29 Oct 2026Self-reward and self-play
Predict the Reward, Not the World: World Reward Models for ML Agents
The loop: A world reward model learns from real execution results to directly choose the best candidate programs written by an ML agent. This model then replaces slow execution steps in the agent's search, making the agent faster and more accurate at finding optimal programs.
Bypassing intermediate execution predictions in favor of direct reward modeling removes a major source of compounding error in search-based ML agents.
- 0229 Oct 2026Training data
Search-Guided Reasoning Trace Curation for Verilog Generation
The loop: A search framework explores intermediate hardware design plans to curate reasoning traces. These traces are then used as supervised fine-tuning data to improve the Verilog generation model that performs the search.
- 0331 Oct 2026Tools and harness
Show HN: A language model that lives in a boot sector
The loop: train.py performs the same integer arithmetic as the assembly code, which improves the inference engine by enabling bit-perfect debugging and alignment, which makes the training script better at producing weights for that specific hardware constraint.
- 0429 Oct 2026Tools and harness
Video Generation Meets Agent Harness: Agentic Video Generation, Agentic World Modeling and Recursive Self-Improvement
The loop: An agent harness coordinates video models and world models to generate interactive environments. These models then mutually evaluate and improve one another, enhancing the harness's ability to produce coherent agentic video generation.
- 0529 Oct 2026Tools and harness
WebAnything: From the Wild Web to Scalable Agent Environments
The loop: An automated framework explores and reconstructs real websites into reusable environments. These environments are then used to train and evaluate web agents, improving their ability to navigate complex web tasks.
- 0626 Oct 2026AI R&D evals
PAPERBENCHX: Can AI Reproduce Science Across Disciplines?
The loop: AI agents attempt to reproduce scientific papers by operating domain-specific software and diagnosing computations. The resulting benchmark measures their capability, providing a target to train more capable autonomous research agents.
- 0727 Oct 2026Kernels and compilers
MUSE: Multi-Level Memory with Selective Runtime Injection for Scientific Coding Agents
The loop: A coding agent builds a multi-level memory from its own historical trajectories. It then selectively injects these past implementation lessons at runtime to improve its performance on kernel optimization and software repair.
- 0827 Oct 2026Inference efficiency
Show HN: A language model that lives in a boot sector
The loop: A training script performs the exact integer arithmetic operations as the target assembly code. This alignment enables bit-perfect debugging of the inference engine, making the script better at producing valid weights for the constrained boot sector hardware.
Loop index
All 74 entries →
- Gemini models propose code changes
- Evaluators keep best scoring versions
- Finds better Gemini training kernels
- Lowers cost to train next Gemini models
- repeat
- Coding agent edits its own code
- Scored child agents enter the archive
- Archived agents make later self modifications
- repeat
- RL agent lays out TPU blocks
- Better layouts make AI hardware cheaper
- Agent pre-trains on earlier chip generations
- repeat
- Coding agent edits LLM training script
- Runs a 5-minute training job
- Keeps change if validation improves
- Changes transferred to larger models
- repeat
- Research agent writes harness mutation blueprints
- Coding agent implements harness module changes
- Winning harness becomes the new champion
- New champion harness runs further generations
- repeat
- Aligned model generates training data
- Reward model judges and filters it
- Data trains next intermediate instruct model
- Next model generates better alignment data
- repeat
- Model generates candidate responses to prompts
- Model scores them itself as a judge
- Scores become preference pairs for DPO
- Trained model becomes next generator and judge
- repeat
- Claude Code writes its own code changes
- Changes become next Claude Code versions
- Release becomes harness for next development round
- repeat
- LLM agents write and benchmark kernels
- Kernels speed up AI model training
- Session data post-trains smaller kernel models
- repeat
- Coding agent writes fine-tuning code
- Agent post-trains a base LLM
- Trained model scored on target benchmark
- One turn of model training another
- repeat
- Learned optimizers update other learned optimizers
- Training keeps best performing optimizer parameters
- Better optimizers train each other faster
- repeat
- LLM reads reasoning model thoughts
- LLM flags reward hacking behavior
- Signal feeds back into training reward
- Trains more aligned capable agents
- repeat
Support for loop research
How to apply →Grants
Funding for projects that put AI in the loop of making AI better. Rolling, low paperwork, sized to the work.
About grants →Fellowships
Funded time, compute and mentorship for researchers who want to go deep on one loop.
About fellowships →Compute
GPU time and credits for loop experiments, which almost always have to run more than once.
About compute →Have a loop worth building?
Tell us what the AI improves and how the improvement comes back. Applications for grants, fellowships and compute all start on the contact page.