Master's graduate from Beijing Institute of Technology, published four top‑journal papers, previously worked as a developer at ByteDance and Alibaba. Currently researching large models at a major state‑owned enterprise. Committed to sharing concise, practical AI large‑model development experience, believing that AI large models will become as essential as PCs in the future. Let's start experimenting now!
The article traces the origin of Loop Engineering, defines it as an autonomous loop system for AI agents, outlines its evolution from Prompt to Context to Harness Engineering, and explains the two core steps—automated start and verification—along with practical implementation details.
This guide walks developers through six curated GitHub repositories that enable advanced multi‑agent usage of Claude Code, teach the fundamentals of building a custom code‑agent from scratch, and provide deep source‑code analysis for a complete understanding of AI‑powered programming assistants.
The article outlines a progressive learning path for AI agents, recommending six GitHub projects—from a beginner-friendly tutorial to production‑grade frameworks—detailing each project's purpose, difficulty, key takeaways, and suitable audience, helping programmers transition from users to builders.
This article walks through practical large‑model evaluation using the EvalScope platform, covering dataset‑based testing, multi‑dataset aggregation, custom data creation, the BLEU and ROUGE metrics, and how to employ a judge LLM for automated, quantifiable scoring.
This guide explains why systematic evaluation is essential for large models, outlines three core evaluation approaches—human assessment, benchmark‑dataset testing, and automated judge models—introduces the most widely used benchmark suites, and shows how to use the open‑source EvalScope framework and prompt‑design techniques to conduct reliable model assessments.
This article provides a step‑by‑step guide to using Agent Skills in LangChain DeepAgents, covering the Skills directory structure, the four engineering steps (discovery, system‑prompt injection, progressive loading, execution), and two practical examples—a simple skill lookup and a complex docx‑processing skill—complete with code snippets and troubleshooting tips.
This guide walks you through interpreting training logs and loss curves, diagnosing common issues such as oscillation, under‑fitting, and over‑fitting, and applying concrete adjustments to learning rate, LoRA settings, batch size, and epochs, with scenario‑specific strategies to turn a novice into a tuning expert.
The article explains Harness Engineering, the emerging concept that moves AI agents from simple question answering to reliable task execution by adding constraints, orchestration, observation, and recovery mechanisms, and shows how it builds on Prompt and Context Engineering through layered architecture and real‑world examples from OpenAI and Anthropic.
This article explains DeepAgents SubAgent mechanisms, showing how context isolation and task division improve complex agent workflows, details two creation methods (dictionary‑based and compiled), demonstrates a search‑and‑report demo, and outlines suitable and unsuitable scenarios with practical code examples.
This article walks readers through the three major training methods—full‑parameter, LoRA, and QLoRA—explaining their memory costs, data requirements, and trade‑offs, then provides a line‑by‑line breakdown of LLaMAFactory configuration files, hyper‑parameter tuning guidelines, and the process for merging LoRA adapters into a deployable model.
