#runtime engineering
1513 related articles
·2 min
Daedalus: A Deep Dive into the Ollama-Based Local AI Engineering Runtime
Daedalus is an open-source local AI engineering runtime built on Ollama, covering architecture, debugging, and security. Zero token costs, full privacy, integrates with Claude Code and OpenCode.
Read more →
·4 min
Why Do AI Agents Crash the Moment They Go Live? A Practical Guide to Harness Engineering for Runtime
A deep dive into the Claude Code source code, systematically analyzing the five-layer Harness Engineering architecture: environment, tool, control, memory, and evaluation. Build a stable runtime system for production AI Agents.
Read more →
·4 min
vLLM Deployment & Unsloth Fine-Tuning in Practice: A Complete Guide to LLM Inference Services
Complete guide to vLLM inference deployment and Unsloth fine-tuning, covering CLI deployment, Python integration, AutoDL cloud setup, and ModelScope acceleration with DeepSeek-OCR as a practical example.
Read more →
·2 min
Quickly Understanding Legacy Project Architecture with OpenSpec Explore
Learn how to use OpenSpec explore in VS Code with GitHub Copilot to automatically analyze legacy project architecture, tech stack, and core features for rapid codebase understanding.
Read more →
·4 min
Qwen 3.8 27B Hands-On: Running a Long-Horizon Coding Agent on a Single GPU
Qwen 3.8 27B local deployment hands-on: 4-bit quantization on a 24GB GPU, SGLang inference pitfalls, coding and long-horizon task testing. SWE-bench Pro surpasses Claude Opus—local long-horizon coding becomes reality.
Read more →
·3 min
The Dilemma and Way Forward for Formal Verification: Lessons from 50 Years of Debate
Revisiting the 1979 DeMillo critique of formal verification: examining whether modern tools like Coq, TLA+, and Lean solve fundamental issues of specification correctness and social processes.
Read more →
·4 min
Agentic Engineering: How AI Agents Are Reshaping Physics Simulation and Robotics Development
Deep dive into the agentic engineering paradigm from NVIDIA's SIGGRAPH demo—from vibe coding to controlled workflows, and how Omniverse libraries empower AI Agents for physics simulation and robotics.
Read more →
·3 min
AI Testing Implementation Guide: Three Major Pain Points and Cost-Effective Solution Selection
Analyze three real pain points of AI in software testing — output randomness, Token costs, and execution efficiency — with a detailed guide to the "AI generation + code execution" approach for optimal cost-effectiveness.
Read more →
·3 min
Langfuse Practical Guide: Core Capabilities and Boundaries of an AI Agent Observability Platform
A deep dive into the open-source LLMOps platform Langfuse: its core positioning, agent trace tracking, prompt version management, token cost analysis, evaluation feedback, and capability boundaries for production AI observability.
Read more →
·3 min
AI-Memory: Building a Cross-Tool Long-Term Memory System for Coding AIs
AI-Memory is a Rust-based open-source project providing long-term memory for Claude Code, Cursor, Aider and other Agent coding CLIs, enabling seamless handoff between vendors.
Read more →
·3 min
Ballet: Codifying AI Workflows to Deliver Deterministic Results Every Time
Ballet converts natural language workflows into deterministic code execution, with audit logs, one-click rollback, simulation mode, and enterprise features to solve AI Agent reliability challenges.
Read more →
·4 min
Go Microservices in Practice: Detailed Architecture for E-Commerce, AI Agent, and IM System Integration
Deep dive into integrating e-commerce, AI Agent, and IM systems under Go microservices architecture, covering unified auth, gRPC, componentized Agent engines, and group chat bots.
Read more →
·3 min
Beginner's Guide to Agents: A Complete Breakdown of MCP, CLI, and Skills Core Concepts
Plain-language guide to AI Agent core concepts: from LLM brains and Chain of Thought reasoning to MCP protocol, CLI tool calling, ReAct loops, Plugins, and Skills—understand how agents work.
Read more →
·2 min
Go + Electron + MCP: Building an AI-Collaborative Task Board
A task board tool designed for AI collaborative development, using Go + Electron + MCP to provide one dataset with two views—visual kanban for humans and structured interfaces for AI Agents.
Read more →
·2 min
DSH Modern Theme: An In-Depth Analysis of the Modern Interface Plugin for DeepSeek Harness
In-depth analysis of the DSH Modern Theme plugin for DeepSeek Harness, featuring soft gray canvas design, five brand color schemes, archived session management, and cross-platform installation.
Read more →
·3 min
RunTrace: A Local-First CLI Tool for ML Experiment Reproducibility
RunTrace is a lightweight open-source CLI tool that saves reproducibility context for ML experiments by recording Git status, Python environment, GPU info, and config files. Local-first with zero server dependencies.
Read more →
·4 min
AI-Driven Testing in Practice: A Complete Guide to Building an Agent Testing Workbench
A complete guide to building an AI-driven testing workbench with five-layer architecture, covering Claude Code agent client setup, DeepSeek model integration, and Node.js environment configuration.
Read more →
·3 min
Hugging Face Reproduces 2,200 ICML Papers: AI Research Reproducibility Crisis Fully Exposed
Hugging Face attempted to reproduce 2,200 ICML papers, exposing the machine learning reproducibility crisis. Analysis of code gaps, fragile dependencies, and compute barriers with paths forward.
Read more →
·3 min
Building an LLM Inference Engine in Rust That Rivals Llama.cpp
Exploring the technical path to building an LLM inference engine in pure Rust that rivals Llama.cpp, analyzing Rust's advantages and challenges in memory safety, SIMD optimization, and GPU backends.
Read more →
·3 min
Sandcastle: An Open-Source Sandbox Orchestration Tool for Running AI Coding Agents Unattended
Sandcastle is an open-source TypeScript library that lets AI coding Agents like Claude Code run unattended in parallel via Docker sandbox isolation, with full workflow orchestration from GitHub Issues.
Read more →