agent-training

Agentic RL 中文零基础教程（33 章）：从概念到 GRPO 实战，含 TRL 最小可跑示例；26–33 章附一套可运行的三方判别模型实证工程（encoder vs LLM-LoRA vs 规则基线）。第 25 章讲清 Jev / TypeSafe System One 与 RL 的能力边界 | Chinese Agentic RL tutorial (33 chapters) + a reproducible discriminative-model benchmark

A curated list of awesome resources about reward construction for AI agents. This repository covers cutting-edge research, and practical guides on defining and collecting rewards to build more intelligent and aligned AI agents.

[NeurIPS 2026] FedAgent: A Library for Decentralized Agent Learning. 🏆 Best Paper Award at the AAAI'26 TrustAgent Workshop, 🏆 Outstanding Paper Award in the AAAI'26 PerFM Workshop.

Official Code for AutoLLMResearch: Training Research Agents for Automating LLM Experiment Configuration — Learning from Cheap, Optimizing Expensive

Self-Mutating Agent Gym

A curated list of agentic environment synthesis, evolution, quality & scaling — built on three surveys (AEE, Environment Scaling, ACE lens)

Turn any real software into a replayable RL environment for training AI agents — deterministic replay, verifiable rewards, TRL & verifiers adapters.

Synthetic environments for training robust tool-use agents via trajectory synthesis

AI 个人记忆训练库：让 AI 记住你的沟通习惯与环境事实，越用越懂你（两级记忆架构，Claude Code 适配）

14 Principles. 23 Real Corrections. A systematic methodology for training AI agents through correction loops.

The Vercel for Agent Training - Train production-ready AI agents with 95% tool reliability

Implicit is the experience layer for AI agents. For large agent training/evaluation environments, address experiences and materialize required state on demand. Python package: implicit-ai.

Audit synthetic agent-training tasks for environment feasibility, verifier quality, evidence gaps, and batch coverage.

AI schooling: experienced agents train newly hatched ones, then examine them cold to graduate them. The coop is the local RAPP neighborhood - several twins (human and AI), one world, no collisions. Pattern dedicated to the public domain.

Kinesift is the observation module entry point; the main project is Truquence, available on PyPI as truquence.

The Flight Simulator for Production AI Agents — Generate high-quality synthetic trajectories for training reliable SRE, DevOps, and infra agents.

Deep reinforcement learning project comparing DQN variants including baseline DQN, Double DQN, curriculum learning, and reward shaping.

Neurochemical behavior training for AI agents — PentaDrive model with 5 drives, 3 phases, MCP server, and structured training modules

Local macOS recorder for voluntary computer-use demonstrations and agent-training datasets. Timestamped inputs, cursor trails, and read-only MCP. Experimental.

Latest advances in agentic kernel optimization.