Developers
CLAUDE.md and skills, subagents and agent teams, loop engineering, keeping token costs down, verifying AI-written code, and the skills engineers need as writing code gets cheaper.

Build These 5 Projects to Master LLM Engineering in 2027
Master LLM engineering by building 5 production systems: Hybrid RAG, multi-agent workflows, AI gateways, automated evaluation suites, and fine-tuned SLMs.
Tuan Tran Van
The New Software Lifecycle: Vibe Coding to Agentic Engineering
The new software lifecycle shifts engineering focus from writing syntax to context management, harness architecture, and trajectory verification.
Tuan Tran Van
What is Grok Bot? The AI Teammates That Actually Finish the Work
Grok Bot provides persistent cloud microVMs operating as always-on AI teammates to automate complex, multi-step digital workflows end-to-end.
Tuan Tran Van
Practical Loop Engineering: Don't Delegate Your Judgment
Loop engineering replaces manual turn-by-turn prompting with autonomous, self-correcting cycles that execute, evaluate, and verify software tasks.
Tuan Tran Van- Must read

Inside the Claude Code Architecture: Designing an Agentic Harness
The Claude Code architecture relies on autonomous loops, minimal tool abstractions, strict permission modes, and aggressive context window management.
Tuan Tran Van - Must read

How the Claude Code Team Does Code Review
The Claude Code team handles code review at agentic scale with pre-PR verification loops, multi-agent checks, and risk-tiered human approvals.
Tuan Tran Van - Must read

Sandboxes Aren't Optional: 4 Ways to Isolate Coding Agents
You must isolate coding agents using multi-layered runtime boundaries to protect host toolchains and prevent credential leakage.
Tuan Tran Van - Must read

Reverse Engineer Anything: What the Viral REA MCP Actually Does
The REA MCP bridges AI coding agents to tools like Ghidra and Hopper, letting developers decompile, analyze, and rebuild app features locally.
Tuan Tran Van 
Choosing the Right Claude Model: Haiku, Sonnet, Opus, or Fable
Optimize performance and token expenditure when choosing Claude model tiers by balancing task complexity against cost, speed, and depth.
Tuan Tran Van- Must read

Own the Outer Loop: Let Agents Write the Code, Keep the Merge Call
Loop engineering requires humans to own the outer loop, boundary control, and production verdicts to stop unverified AI code from breaking systems.
Tuan Tran Van - Must read

The State of the Tech Industry in 2026: Code Is Cheap, Judgment Isn't
The tech industry in 2026 has shifted to parallel agent orchestration, turning raw code into a disposable cache verified by production rigor.
Tuan Tran Van 
Graph Engineering: The Karpathy Loop and Anthropic's Playbook
Graph engineering transitions autonomous agent swarms from volatile single-agent iteration loops into durable, graph-grounded execution systems.
Tuan Tran Van- Must read

From Loop Engineering to Graph Engineering: 4 Steps to Durable Agents
Evolve AI agents from loop engineering to graph engineering by externalizing shared state, context windows, and execution history into durable memory.
Tuan Tran Van - Must read

Why Code Verification Matters More Than Ever in the Age of AI
Deploying AI-generated software requires multi-layered code verification to prevent systemic failures and maintain long-term code quality.
Tuan Tran Van - Must read

Audit Your Agent Files: Pruning CLAUDE.md and AGENTS.md Effectively
Regularly prune CLAUDE.md and AGENTS.md files to eliminate context bloat, reduce token costs, and improve AI coding agent adherence.
Tuan Tran Van - Must read

Brownfield Agentic Engineering: Running AI Agents in Legacy Codebases
Practicing brownfield agentic engineering requires mapping codebase risk zones, pinning legacy behaviors with tests, and establishing strict harnesses.
Tuan Tran Van - Must read

Human Judgment Doesn't Leave the Software Factory. It Relocates.
Automated agent fleets do not eliminate engineering liability, as human judgment relocates upstream to intent, control harnesses, and verification boundaries.
Tuan Tran Van - Must read

Agentic Skill Decay: Why Completing Tasks Is Not Doing the Reps
Delegating coding tasks to AI causes agentic skill decay. Learn why active reps and verification are critical to preserve long-term technical mastery.
Tuan Tran Van - Must read

Multi-Agent Architectures, Clearly Explained
Multi-agent architectures resolve context overflow and parallelism constraints using orchestrators, pipelines, and hierarchical supervisors in production.
Tuan Tran Van 
What is Jev? The System One Model for Code Decisions
Jev is a System One model from TypeSafe AI designed for fast, structured decisions in software applications, offering calibrated, typed outputs at low latency.
Tuan Tran Van- Must read

Claude Code Best Practices Every Developer Should Know
Master Claude Code best practices to optimize agentic workflows, manage context rot, and leverage auto mode for safer, more efficient software engineering.
Tuan Tran Van - Must read

Agentic Software Factory: The Future of Software Engineering
The agentic software factory uses autonomous agents and standardized tooling to industrialize software production and scale velocity through a managed loop.
Tuan Tran Van 
What Is RAD? Why It Matters in the Age of AI Coding
Rapid Application Development (RAD) is resurfacing as a framework for agentic AI workflows, built on iterative prototyping and spec-driven testing.
Tuan Tran Van- Must read

How to Stop Hitting Claude Code Usage Limits
The complete guide to Claude Code usage limits: context hygiene, model routing, cache hits, hook enforcement, subagents, and offloading I/O to cheaper models.
Tuan Tran Van - Must read

Why You Still Need to Learn the Fundamentals in the Age of AI
You need to learn the fundamentals to audit AI-generated code. Without that grounding, the costs are security holes and skill atrophy.
Tuan Tran Van - Must read

A Complete AI Coding Workflow with mattpocock/skills
Automate your AI coding workflow from initial design to final review using mattpocock/skills for a structured, test-driven engineering pipeline.
Tuan Tran Van - Must read

How to Build Your Own Coding Agent
Build a functional coding agent with Python or Go to automate codebase navigation, surgical file editing, and guarded shell command execution in a loop.
Tuan Tran Van - Must read

The Best Setup for the .claude Folder for Maximum Efficiency
The .claude folder gives Claude Code a hierarchical config that optimizes token usage through modular rules, skills, and deterministic hooks.
Tuan Tran Van - Must read

Why You Should Delete Claude Code's Auto Memory (memory.md)
Disable the memory.md system in Claude Code to eliminate context bloat, prevent state drift between worktrees, and ensure deterministic agent behavior.
Tuan Tran Van 
Ultrathink, Ultraplan, Ultracode, Ultrareview in Claude Code
Claude Code's four ultra commands: Ultrathink reasoning, Ultraplan cloud drafting, Ultracode parallel agents, and verified Ultrareview bug detection.
Tuan Tran Van- Must read

CLAUDE.md vs AGENTS.md vs SKILL.md: How to Combine All Three
How CLAUDE.md vs AGENTS.md vs SKILL.md split by loading model — decide where every instruction goes, protect your context budget, and cut agent runtime.
Tuan Tran Van - Must read

Free Claude Code Skills That Fix 90% of That "AI-Generated" UI Look
Replace generic AI-generated UI defaults with opinionated engineering skills to implement distinctive visual systems and deterministic design rules.
Tuan Tran Van - Must read

What is /compact? The command that fights context rot in Claude Code
Manage context rot with the /compact command to keep Claude Code session health by replacing verbatim history with structured, high-quality summaries.
Tuan Tran Van - Must read

How Test-Driven Development Amplifies AI Success
Test-Driven Development gives AI agents deterministic guardrails, cutting bugs by up to 80% and keeping non-deterministic output stable.
Tuan Tran Van 
Tips & Tricks to Cut Claude Fable 5 Costs by 80%
Reduce your Claude Fable 5 cost by up to 80% using effort tuning, prompt caching, the 10-80-10 routing system, and image-based context compression.
Tuan Tran Van
How Anthropic's Teams Use Claude Code in Real Work
See how Anthropic teams use Claude Code to accelerate debugging, automate testing, and let non-technical departments build their own custom software tools.
Tuan Tran Van- Must read

How Does Memory Work in Claude Code?
Optimize Claude Code memory using CLAUDE.md and auto-memory systems to keep persistent project context and high instruction adherence across sessions.
Tuan Tran Van - Must read

Understanding Spec-Driven Development
Spec-driven development makes a structured specification the primary source of truth for AI coding agents, replacing vibe coding with a verifiable contract.
Tuan Tran Van - Must read

The Best Claude Code Skills to Make Claude Work Like a Senior Engineer
A lean, audited checklist of the Claude Code skills worth installing — the ones that make Claude plan, test, review, and ship like a senior engineer.
Tuan Tran Van - Must read

The GitHub Repos and Websites That Help You Master Claude Code
A hand-picked roundup of the GitHub repos and websites that help you master Claude Code and work better with AI.
Tuan Tran Van - Must read

Claude Code's /goal Command: Engineering Autonomous Cycles
The /goal command in Claude Code enables autonomous execution of long-running development tasks using an evaluator-loop and persistent completion conditions.
Tuan Tran Van 
The /loop Command in Claude Code: When and How to Use It
Learn how to use the loop command in Claude Code to automate in-session tasks. Master cron scheduling, jitter logic, and session-scoped automation.
Tuan Tran Van- Must read

Loop Engineering: Building Software With Agent Loops
Learn loop engineering: the transition from manual prompting to designing autonomous systems that use agents to discover, implement, and verify work.
Tuan Tran Van - Must read

The Ralph Loop: Run AI Agents in a Bash Loop Until They Finish
Achieve autonomous AI coding with the Ralph Loop. Solve context rot, implement PIN systems, and use bash-scripted iterations for AFK development.
Tuan Tran Van - Must read

A Practical Guide to Claude Memory Files
Claude memory files give an agent project-level persistence and steering, using CLAUDE.md and hooks to keep context lean and automate workflows.
Tuan Tran Van - Must read

Harness Engineering: Build the Scaffolding Around an AI Model
Harness engineering builds the deterministic scaffolding, feedback loops, and constraints around an AI model so agents run reliably in production.
Tuan Tran Van - Must read

Why You Should Stop Using /init to Generate Your CLAUDE.md
Stop using /init for your CLAUDE.md. Engineering benchmarks show auto-generated files raise costs by 20% and dilute the agent's context with noise.
Tuan Tran Van - Must read

AGENTS.md: What Works, What Costs, and Best Practices
What the data says about AGENTS.md context files: the 4% gain, the 20% cost tax, and how to write one that helps repository-level agents.
Tuan Tran Van - Must read

What CLAUDE.md Is and How to Write an Effective CLAUDE.md File
Configure CLAUDE.md to eliminate AI amnesia. This guide covers context engineering, file hierarchy, and behavioral rules for efficient Claude Code sessions.
Tuan Tran Van - Must read

Claude Code Commands & Shortcuts Cheatsheet
Quick-reference tables explaining every useful Claude Code slash command, keyboard shortcut, and hidden feature.
Tuan Tran Van - Must read

What Is Agentic Engineering? A Guide for Engineers
Agentic engineering is a disciplined practice where AI agents handle implementation while humans oversee architecture and validation to ensure reliability.
Tuan Tran Van 
Context Engineering vs Prompt Engineering
Master context engineering vs prompt engineering to build resilient AI agents using RAG, memory management, and advanced orchestrator architectures.
Tuan Tran Van- Must read

Everything you need to know about prompt engineering
The core techniques and developer workflows of prompt engineering: build reliable AI systems, debug code faster, and scaffold new features.
Tuan Tran Van 
Dynamic Workflows, Sub-agents, or Agent Teams: Which to Use
Compare dynamic workflows, sub-agents, and agent teams in Claude Code. Optimize for token efficiency and scale using this technical decision framework.
Tuan Tran Van
Mastering Claude Agent Teams
Learn how to orchestrate Claude agent teams to scale development with parallel subagents, a shared task list, and direct inter-agent communication.
Tuan Tran Van- Must read

Mastering Subagents in Claude Code
Use subagents in Claude Code to isolate context and run work in parallel, so the main thread keeps its reasoning quality as a task grows.
Tuan Tran Van - Must read

A Guide to Context Engineering
Context engineering is the systems discipline for managing an LLM's information state to prevent context rot and "Lost in the Middle" performance drops.
Tuan Tran Van 
Mastering Vibe Coding: Best Practices
Learn the engineering rigor behind vibe coding: planning, context management, and code review for production-grade AI software development.
Tuan Tran Van- Must read

Software Engineer in the Age of AI
The role of a software engineer in the age of AI is shifting from writing syntax to designing robust architectures and managing complex system failures.
Tuan Tran Van - Must read

What is vibe coding? Building apps with AI and systems thinking
Vibe coding lets creators use natural language and agentic AI to orchestrate, refine, and deploy production-grade software — fast, but with real trade-offs.
Tuan Tran Van - Must read

The Evolution of Vibe Coding: From Trend to Discipline
Vibe coding raced from viral prototype trick to production liability, and teams are now shifting to agentic engineering where humans own the architecture.
Tuan Tran Van