#LLM pipeline reliability
362 related articles
Tutorials·3 min
Fully Automated Invoice Reimbursement with Local AI Agent: From Process Breakdown to Deployment
How to build a fully automated invoice reimbursement system with local AI Agents, covering OCR, info extraction, and form generation with MinerU+Qwen3+Qianwen Po.
Read more →
Deep Dives·3 min
MCP (Model Context Protocol): Solving the Three Core Pain Points of AI Tool Calling
Deep dive into how MCP (Model Context Protocol) solves three core pain points of Tool Calling: verbose descriptions, unstable invocations, and lack of unified standards.
Read more →
Product Reviews·3 min
6 Developer Tools Roundup: AI Coding Cost Tracking, Agent Virtual File Systems, and More
Roundup of 6 developer tools: CodeBurn for AI coding token cost tracking, Mirage virtual file system for Agents, Boring SSH tunnel manager, PeerTrace file tree renderer, DataTab font-based data visualization, and Flu TypeScript Agent framework.
Read more →
Deep Dives
The Complete Guide to OpenTelemetry: T…
·4 min
The Complete Guide to OpenTelemetry: The Three Pillars of Observability & DevOps in Practice
Deep dive into OpenTelemetry's core architecture, the three pillars of observability (logs, metrics, traces), with a real-world food delivery app crash debugging case showing how distributed tracing locates microservice bottlenecks and how AI is revolutionizing DevOps monitoring.
Read more →
Product Reviews·4 min
Generic Agent: An Open-Source AI Agent That Uses Only One-Tenth the Tokens of Comparable Products
Generic Agent is an open-source AI Agent that reduces token consumption by 90% through minimalist tool design, four-layer memory hierarchy, and experience reuse. Supports computer operation, browser automation, Feishu integration, and more.
Read more →
Research·3 min
SciMDR: How a 7B Small Model Rivals GPT-5 in Scientific Reasoning
Yale and other institutions introduce SciMDR, a two-stage data synthesis pipeline enabling a 7B model to match GPT-5 level performance in scientific literature comprehension.
Read more →
Deep Dives·2 min
RAG System End-to-End Breakdown: From Vector Indexing to Production Optimization
A deep dive into the complete RAG pipeline — covering vector embeddings, document chunking, retrieval and reranking, plus three production optimization techniques for building accurate enterprise AI knowledge base applications.
Read more →
Tutorials·4 min
Building a WeChat Mini Program with Cursor: A Practical Guide to AI Programming Pitfalls and Technical Decisions
Real-world experience building a WeChat Mini Program with Cursor, covering product design, technical pitfalls, and creative LLM API usage — a practical guide for AI programming beginners.
Read more →
Product Reviews·3 min
WhichLLM: One Command to Find the Best Local LLM for Your Hardware
WhichLLM is an open-source tool that auto-detects your hardware and recommends the best local LLM using real benchmark data. Simulate GPUs, filter fake benchmarks, and start chatting in one command.
Read more →
Research·4 min
Deep Dive into Claude Code's Open-Source Architecture: The Design Philosophy Behind 510,000 Lines of Code
Deep analysis of Claude Code's open-source architecture: dual-loop design, 7-step tool pipeline, 4-layer token compression, memory systems, and multi-agent collaboration patterns.
Read more →
Tutorials·3 min
Learn AI Agents in 30 Days: A Four-Stage Learning Roadmap from Zero to Production
A systematic breakdown of the AI Agent learning roadmap covering core architecture, ReAct/CoT paradigms, multi-agent collaboration, and Prompt optimization across four stages with quality resource recommendations.
Read more →
Tutorials·4 min
Semi-AI Mode: A Pragmatic Approach to API Automation Testing Frameworks
Explore the semi-AI approach to API automation testing: why pure AI fails, framework design principles, technology choices, and clear human-AI division of labor for practical implementation.
Read more →
Tutorials·3 min
OpenClaw + Claude Code Token-Saving Guide: Zero-Polling with Hooks Callbacks
How to solve high Token consumption when OpenClaw calls Claude Code. Achieve zero-polling with Stop Hook and Session End dual callbacks, combined with Agent Teams for fully automated dev workflows.
Read more →
Tutorials·2 min
The Complete Evolution of a Quantitative Trading System: From Multi-User Isolation to AI Trading Agent in Practice
A detailed walkthrough of a quant trading system's evolution: multi-user isolation, audit logging, strategy analysis, trading experience system, AI Agent decision-making, and LLM gateway design.
Read more →
Tutorials·3 min
Can You Really Transition to AI/LLM in Three Months? A Deep Dive into the Learning Roadmap
Deep dive into a popular 3-month AI/LLM transition roadmap: from Python basics and Prompt engineering to LangChain, RAG, Agents, and hands-on projects, with realistic time estimates and pitfall warnings.
Read more →
Tutorials
n8n + Claude Code: Build a Fully Autom…
·3 min
n8n + Claude Code: Build a Fully Automated AI Coding Assistant — Send a Slack Message, Get a PR Automatically
Learn how to build an automated workflow with n8n and Claude Code that turns Slack messages into GitHub Pull Requests — from setup to scripts to live demo.
Read more →
Tutorials·3 min
Vibe Engineering in Practice: AI Evolves from Code Generator to Programming Teammate
A deep dive into Vibe Engineering principles: context engineering, sub-agent collaboration, autonomous testing loops, plus OpenAI's case study of a 12-hour Kotlin-to-Rust rewrite.
Read more →
Tech Frontiers·2 min
SGLang v0.5.12.post1 Released: DeepSeek V4 Stability Fixes and Blackwell Adaptation
SGLang v0.5.12.post1 stability patch details: 12 critical fixes covering DeepSeek V4 garbled text and crashes, NIXL PD disaggregated inference logic, Blackwell B300 adaptation, and cold start optimization.
Read more →
Tech Frontiers·2 min
Step 3.7 Flash: Deep Dive into the 198B Sparse MoE Multimodal Model
Deep dive into StepFun AI's Step 3.7 Flash, a 198B sparse MoE vision-language model with 256K context and 3-level reasoning, excelling in multimodal understanding, AI coding, and Agent tool orchestration.
Read more →
Tech Frontiers·1 min
Cloudflare Contributes Critical KV Cache and Mooncake Fixes to SGLang
Cloudflare contributes decode KV cache offload and Mooncake recovery fixes to SGLang, resolving garbled output under high concurrency for Kimi K2.6 and enabling automatic fault recovery in distributed inference.
Read more →