SL#52 - The Debt AI Is Building Isn't In Your Code
Refactoring used to be the antidote to messy code. It can't reach the kind of debt your team is now accumulating, because the debt has moved somewhere refactoring can't go.
SL#51 - Your Coding Agent Didn't Solve SWE-bench. It Cheated.
A team at Berkeley built a scanner that scored near-perfect on eight of the most-cited AI agent benchmarks without solving a single task. The leaderboard you used to pick your model is measuring the wrong thing.
SL#50 - Build a Claude Sub-Agent That Reviews Your PRs Against a Team Style Guide
A 200-line TypeScript build using the Claude Agent SDK. The non-obvious bit: one sub-agent per file beats one big prompt, and it costs less. Plan for 90 minutes if you have used the Agent SDK before, 3 hours if this is your first time.
SL#49 - The Eight-Hour Outage After a Seven-Minute Fix
Railway's Google Cloud account was suspended at 22:20 UTC and reinstated at 22:29. The platform was still down at 06:14 the next morning. The lesson is bigger than Railway.
SL#46 - Your LLM Agent Is Drowning in Its Own Context Window
Context windows just hit two million tokens. So why are 5% of production AI requests still failing? Because the industry confused having more space with knowing what to put in it.
SL#44 - Building Agentic RAG Systems: Architecture, Reasoning Loops, and Production Considerations
The transition from simple LLM wrappers to AI Agents represents the next frontier in software engineering. While traditional Retrieval-Augmented Generation (RAG) improved LLM accuracy, Agentic RAG introduces a reasoning layer that allows the system to autonomously decide how to use data to solve a problem.
SL#43 - Beyond the Chatbox: How MCP Turns LLMs into Autonomous Operators
The Model Context Protocol (MCP) dismantles the "data silos" of modern AI, providing a standardized bridge for LLMs to move beyond conversation and into direct, real-world execution.
SL#42 - From Passive LLMs to Autonomous Agents: The Evolution of AI Workflows
The field of Artificial Intelligence is rapidly evolving from simple text generation to autonomous problem-solving. To understand where the industry is heading, technical professionals must distinguish between three distinct levels of AI implementation: Passive LLMs, AI Workflows, and Autonomous AI Agents.
