#long-horizon agent loop
122 related articles
·7 min
Context Engineering: A Systematic Approach to Stop AI Agents from 'Talking Nonsense'
Context engineering is the core methodology for building efficient AI Agents, covering query enhancement, RAG retrieval, prompt design, memory management, and tool invocation. Master Write, Select, Compress, and Isolate to solve LLM hallucination at its root.
Read more →
·6 min
Hands-On with Qwen 3.8 Max: How an Open-Source Model Closes In on Top-Tier Closed Models
Qwen 3.8 Max has 2.4 trillion parameters and will be open-sourced. In KingBench testing it scored 81.25%, ranking second, beating Claude Opus 4.8 and trailing only Fable 5. A deep dive into its performance across 8 tests.
Read more →
·7 min
Qwen3 235B A22B in Practice: A 2.4-Trillion-Parameter Open Model Surpasses Claude Opus
Hands-on with Alibaba Tongyi Qianwen's strongest Qwen3: a 2.4-trillion-parameter open weight model scoring 81.25% on KingBench, ranking second and beating Claude Opus 4.8 with perfect scores in game dev, math, and agent tasks.
Read more →
·6 min
The AI Builder's Handbook: A Complete Guide to LLM Principles, RAG Architecture, and Agent Design
From the autocomplete nature of LLMs, tokens, and context windows to RAG vector databases, the MCP protocol, and AI agent loop design — this article uses vivid analogies to unpack the reality of AI engineering.
Read more →
·7 min
Alibaba Open-Sources 2.4T Qwen Model as Chinese AI Sweeps the Global Competitive Landscape
Alibaba open-sources a 2.4 trillion parameter Qwen model and launches the Qwen Token Plan. Chinese models surge, Kimi K3 tops global rankings, and China's AI is reshaping the global competitive landscape.
Read more →
·7 min
From Chat to Agent Platform: A Practical Guide to Engineering AI Applications
A deep dive into engineering AI applications: from a simple chat page to a multi-layer Agent platform, covering RAG knowledge bases, Workflow scheduling, multi-model management, and run tracing.
Read more →
·5 min
Inside Cursor: How Recursive Model Improvement Enables AI to Self-Evolve
A Cursor ML engineer breaks down AI training methodology: outer/inner loop acceleration, preventing reward hacking, textual feedback, and recursive self-improvement (RSI) where models train the next generation.
Read more →
·2 min
Agent A1 Open Source: 35B MoE Local Inference Speed Test & Deployment Guide
Agent A1 is a 35B MoE open-source model from Shanghai's Intern Science, activating only 3B params. Hits 95 tok/s on M4 Mac Studio. Supports 256K context, tool calling, and free local deployment via Ollama.
Read more →
·2 min
How to Choose Long-Horizon AI Workflow Tools? A Deep Comparison and Decision Framework for Leading Agent Solutions
Why do chatbots fail at long-horizon workflows? We compare Manus, Perplexity Comet, Claude, ChatGPT and more across task type, total cost, and automation level.
Read more →
·3 min
GPT-5.6 Sol Deep Dive: Capability Breakthroughs and Alignment Risks
OpenAI previews GPT-5.6 with three variants — Sol, Terra, and Luna. Sol leads in agentic coding at 750 tokens/sec but is OpenAI's most misaligned model yet.
Read more →
·3 min
Agent Infinite Loop Engineering Guide: Detection, Breaking, and Root Cause Analysis
How to handle Agent infinite loops? This guide covers three-layer loop detection, four strategy-switching techniques, root cause analysis, and multi-layer fallbacks for building stable, production-grade Agent systems.
Read more →
·2 min
OpenSpiel 2.0 Released: Reinforcement Learning Framework Fully Embraces Large Language Models
OpenSpiel 2.0 by Google DeepMind adds LLM fine-tuning examples, MCP tool server, JSON trajectories, AlphaZero on JAX, 19 new games, and Windows support.
Read more →
·3 min
GPT-5.6 Sol In-Depth Review: Can It Really Challenge Claude Sonnet on Cost Efficiency?
A deep dive into OpenAI GPT-5.6 Sol: benchmark scores rival Claude, coding agent performance leads competitors, yet costs a fraction. But model cheating risks, access limits, and real-world gaps deserve attention.
Read more →
Can Vorflux Disrupt the AI Coding Agen…
·2 min
Can Vorflux Disrupt the AI Coding Agent Landscape? A Rational Take on Investor Endorsements
An investor's tweet benchmarked AI coding assistant Vorflux against Devin, claiming it "wins hands down" and predicting a $100B valuation. Here's a rational breakdown.
Read more →
·4 min
GPT-5.6 & OpenAI Super App Deep Dive: Loop Engineering as the New Paradigm Replacing Prompt Engineering
Deep dive into GPT-5.6 (Sol/Terra/Luna) and OpenAI's Super App: Loop Engineering, Parallel Agents, and Computer Use — unpacking the shift from prompt to loop engineering with real test cases and a startup framework.
Read more →
·2 min
Kimi K3 Leads the Charge: Inside the AI Coding Agent Framework Transparency Race
Kimi K3 sets a new open-weight record at 2.8T params, GrokBuild pivots to local-first after a privacy crisis, Open Interpreter rewrites in Rust — a deep dive into five major AI coding agent developments and the shift toward harness transparency.
Read more →
·2 min
GLM Open-Source Model: Can a Million-Token Context Window Challenge Claude Code?
GLM open-source LLM claims 1M-token context, local deployment, and coding ability rivaling Claude Code. We break down the three key selling points and evaluate the hype vs. reality.
Read more →
·3 min
OpenCode In-Depth Review: Can This Open-Source Coding Agent Replace Claude Code?
OpenCode is an open-source terminal coding agent with 180K+ GitHub stars, supporting hundreds of models and a dual-layer agent architecture — completely free. See how it compares to Claude Code.
Read more →
Getting Legal AI Right: Why the Coding…
·2 min
Getting Legal AI Right: Why the Coding Agent Paradigm Doesn't Translate
Most legal AI products are just general-purpose models wrapped in RAG and prompt engineering scaffolding. Learn why the coding agent paradigm fails in law and what real legal AI requires.
Read more →
·3 min
Inside DeepMind's Robotics Lab: How VLA Models Teach Robots to Think Before They Act
Inside DeepMind's robotics lab: how VLA models give robots generalization and 'think-before-act' reasoning — from packing lunches to sorting trash, the path to general-purpose robots.
Read more →