Alphabell.
Daily edition

Radar, 19 Nov 2026

Today's radar highlights a shift toward infrastructure and tooling loops, from agents optimizing their own quantization recipes and kernels to a live software factory maintaining its own open-source framework.

PaperarXiv·19 Nov 2026·Self-reward and self-play

SWE-growth: Teacher-Guided Adaptive Self-Play for Software Engineering Agents

The loop: A student coding model attempts repository-level repair tasks, and a teacher model identifies learning opportunities from the failures. These verified outcomes are used as training data for supervised fine-tuning and reinforcement learning, making the student model better at software engineering.

A teacher-guided self-play framework that uses verified outcomes to generate training data for a student coding model, improving its performance on SWE-bench.

loop fit 10/10Songshi Liang, Zixuan Huang, Mingxuan Song, Haoyu Tang, Shusheng Xu, Tongkai...via arXiv
SWE-growth: Teacher-Guided Adaptive Self-Play for Software Engineering Agents
PaperarXiv·19 Nov 2026·AI R&D evals

RSIBench-Infra: Can Agents Optimize Their Own Training and Inference Infrastructure at Production Scale?

The loop: Autonomous agents modify the software stack across engine, kernel, and instruction levels to optimize training and inference workloads. This improves the efficiency of the infrastructure, allowing the same agents to run or train faster in subsequent iterations.

A benchmark of 20 tasks evaluating whether agents can deliver end-to-end performance gains by optimizing production-scale training and inference infrastructure.

loop fit 9/10Hanlin Wu, Yifei Wang, Zhicheng Cai, Huan-ang Gao, Wei-Ying Ma, Ya-Qin Zhang,...via arXiv
RSIBench-Infra: Can Agents Optimize Their Own Training and Inference Infrastructure at Production Scale?
PaperarXiv·19 Nov 2026·Inference efficiency

QDA: Closed-Loop LLM Quantization Design with Agents

The loop: An LLM agent designs and evaluates post-training quantization recipes for large language models. This improves the inference efficiency of those models, making the agent's own execution or its successor's deployment faster and more cost-effective.

A system where an agent autonomously refines quantization recipes for large language models by building and evaluating candidates against reasoning traces.

loop fit 9/10Xingyang Li, Tianlang Zhao, Song Han, Ligeng Zhuvia arXiv
QDA: Closed-Loop LLM Quantization Design with Agents
PaperarXiv·19 Nov 2026·Kernels and compilers

Arbor: How Search Policy Shapes Agent Behavior in System Optimization

The loop: Autonomous agents search for and write kernel-level and framework-level optimizations across an explicit optimization tree. These optimizations speed up the training and inference of AI models, improving the execution of the agents themselves.

A system that provides autonomous agents with an explicit optimization tree to reason over and write kernel-level optimizations for model training and inference.

loop fit 9/10Neha Prakriya, Chaojun Hou, Zheng Gong, Huasha Zhao, Yanyuan Qin, Chris Cai,...via arXiv
Arbor: How Search Policy Shapes Agent Behavior in System Optimization
Blog postOpenAI·18 Nov 2026·Hardware and chips

Advancing chip design with OpenAI models

The loop: OpenAI models optimize circuits and RISC-V core designs for silicon. These hardware improvements are then used in the chips that run the models, accelerating their future execution.

OpenAI reports using its models to optimize circuits and CPU cores for AI-assisted chip design.

loop fit 9/10via OpenAI
Advancing chip design with OpenAI models
ArticleHacker News·18 Nov 2026·Self-modifying agents

Show HN: Mastra Factory, the open-source software factory we use to build Mastra

The loop: Mastra's durable agents operate a software factory that opens and merges pull requests across the project's codebase. This work directly maintains and improves the Mastra framework, which is the very tool used to build and run the factory.

An open-source agentic kanban board that uses durable agents to autonomously open and merge hundreds of pull requests a week to maintain its own underlying framework.

loop fit 9/10codekaratevia Hacker News
Show HN: Mastra Factory, the open-source software factory we use to build Mastra
ProjectGitHub·2 Nov 2026·Tools and harness

lisresearchgroup/math-rsi

The loop: A self-improving agent uses a research loop and verifiable proofs to make mathematical discoveries. These verified proofs expand the agent's memory and proof capabilities, making the harness more effective for further mathematical discovery.

An open-source agent harness for mathematical discovery that uses a shared research loop and verifiable proofs in Lean 4.

loop fit 8/10lisresearchgroupvia GitHub
lisresearchgroup/math-rsi

← 18 Nov 2026 Latest