#AI编程
9638 related articles
AI Agent Enterprise Practical Guide: From Low-Code Platforms to LangChain Framework Implementation
A Bilibili AI Agent enterprise course covering low-code platforms, MCP protocol, and LangChain. Includes 9 job roles and 4 capstone projects for real-world deployment.
How to Quantitatively Evaluate Agent System Engineering Quality? A Six-Dimension Framework Explained
Beyond model capabilities, how do you quantitatively evaluate Agent system engineering quality? This article breaks down a six-dimension framework covering outcome, process, reliability, cost, observability, and security — with Cost per Success as the North Star metric, a four-layer evaluator system, and a release gate flywheel for continuous improvement.
AI Agent Builder: A Beginner's Guide to the Hottest New Role in LLM Deployment
AI Agent is the core of LLM deployment. Learn about the emerging AI Agent Builder role, the SME demand gap, low-code entry paths, and how to go from beginner to delivery.
AI Agent in Practice: Building a Memory-Enabled Intelligent Chat Assistant from Scratch
Step-by-step tutorial: build an AI chat agent with long-term memory using Python, DeepSeek API, and vanilla frontend — master Agent Memory from scratch.
Developer Benchmark: DeepSeek Open-Source Model Now Rivals Claude Opus 4.8
A developer blogger finds DeepSeek V4 Pro rivals Claude Opus 4.8 at a fraction of the cost. Analysis of its open-source harness, BrightData scraping, and SEO use cases.
Andrew Ng's Agentic AI Course Deep Dive: How to Build Truly Useful Agentic AI Applications
A deep dive into Andrew Ng's Agentic AI course: from cutting through the hype to mastering eval-driven development workflows for building truly useful agent applications.
OpenReel: Open-Source Video Creation Studio That Turns One Sentence Into a Short Film, with ComfyUI Handed Off to an Agent
OpenReel open-source video studio by Bilibili creator Ayan: built on DeepSeek-Harness, it hands ComfyUI workflow scheduling to an AI Agent — one prompt generates a complete narration video from brief to final cut.
Overseas Developer Benchmarks DeepSeek: On Par with Opus, Costs Just Pennies
A developer benchmarks DeepSeek V4 Pro against Anthropic Opus 4.8 on real coding tasks, finding near-equal performance at a fraction of the cost — a viable Claude Code alternative.
Bridging DeepSeek's Web Interface into Harness: A Real-World Test of Free Compute for Local Code Editing
A Bilibili creator bridges DeepSeek's free web interface into Harness via a Chrome extension, enabling local file reads and code edits without a paid API. Real-world tests, tool-calling details, and open-source project included.
Revisiting 'Trusting Trust': The Poisoning Risks of Self-Modifying AI Coding Tools
Revisiting Ken Thompson's classic 'Trusting Trust' lecture to explore poisoning risks in self-modifying AI coding tools — covering training data, prompt injection, and self-modification loops.
AI Programming at Xiamen University: A Full Breakdown from Textbook to Teaching
Xiamen University's Lin Ziyu shares how he built an AI programming course — covering three coding eras, Claude Code's production-ready milestone, three methodologies, and a fully free, reproducible curriculum design.
DeepSeek Harness Hands-On: Build Your Own Coding Agent for Free
DeepSeek Harness hits 190K GitHub stars as a free Claude Code alternative. This hands-on review covers installation, plugin models, full traceability, and driving Claude Code as a subprocess.
A DeepSeek Kernel Engineer's Confession: Being Replaced by the AI I Trained
A DeepSeek kernel engineer reveals how AI evolved from a doc-lookup assistant to a kernel master in one year — and why he's accelerating his own replacement.
DeepSeek Harness One-Month Retrospective: Highlights, Risks, and the Road Ahead
DeepSeek Harness one-month retrospective: Plugin ecosystem, Codex Kernel, supply chain risks, talent strategy, and the cloud Agent path ahead.
Open-Source Project Tested: Integrating DeepSeek and Other AI Models into WPS via API
An open-source project uses DSL to integrate DeepSeek and other AI models into WPS, enabling in-document generation and revision. ~8,000 tokens per 2,000 characters. Requires a paid API key.
Vibe Coding in Practice: A Full Breakdown of an Open-Source Live2D AI Desktop Pet Project
A deep dive into an open-source Live2D desktop pet built with Vibe Coding. Electron-based, Windows-ready, with AI chat, TTS voice replies, external messaging, and full character customization.
Dify Self-Hosting Complete Tutorial: Building an AI Application Platform from Scratch
Complete Dify self-hosting tutorial covering Docker Compose deployment, LLM integration, and building your first chat app — perfect for developers and enterprises.
Dify Local Deployment: A Complete Hands-On Guide from VM Setup to AI Agent Building
Complete guide to deploying Dify locally: covers VMware VM setup, Ubuntu 22.04 installation, BT Panel config, Docker deployment, and common network/image troubleshooting tips.
Dify Workflow in Practice: How AI Product Managers Build Business Processes with Low-Code
Based on a Bilibili AI PM course, this article covers Dify workflow concepts, three-layer node structure, deployment, and two real cases — cola pricing and jewelry custom quoting — showing when to use LLM nodes vs. hard rules.
Dify from Beginner to Production: A Complete Learning Roadmap for Building AI Applications
Complete Dify tutorial: Windows Docker deployment, MySQL setup, five app types (Chat/Agent/Workflow), model integration, and publishing — build enterprise AI apps fast.