Author
Sean Robinson
Articles by Sean Robinson (10)
AGENTS.md and CLAUDE.md: Writing Guardrails for AI Coding Agents
Modern AI coding systems read project-level instruction files. Done well, they enforce the patterns you actually want. Done badly, they're prompt-tax with no return.
How Modern LLMs Work: From RNNs to Transformer Attention
Before today's AI, language models were dominated by recurrent neural networks. The shift to Transformer architecture and attention mechanisms changed what context costs — and why your prompt design suddenly matters.
Know-What vs Know-How: The AI Task Taxonomy That Saves You From Disasters
Tasks that require describing WHAT you want behave differently from tasks that require describing HOW you want it done. Getting the category wrong is how you end up with 100k lines of confused code in a day.
Advanced LLM Prompting: Exhaust Categories, Positive Wording, and One Simple Answer
Specific prompting techniques that come from understanding how attention-based models actually choose tokens. Why "do this" beats "don't do that" — and why offering multiple choices hurts quality.
Purposefully Pre-Filling Context: The AI Prompting Pattern You're Not Using
Don't tell the AI what to do — first have it tell YOU how the system works. Then negotiate the change. The two-step context-prefilling pattern that produces consistent results.
RAG, Agents, Memory, and Tool Calls: The AI Infrastructure Stack
Retrieval-augmented generation, agentic systems, persistent memory, structured tool calls — the architectural layer under modern AI deployments. What each one actually does.
The Skipping-Leg-Day Effect: Cognitive Risks of Over-Relying on AI
Using AI assistance for cognitively demanding tasks without staying engaged can cause your own skills to atrophy. Here's the runaway pattern, and how to fight it.
Three Productive Ways to Use Modern LLMs: Tutor, Capable Employee, Recon Tool
There are several patterns that maintain and grow your capabilities while using AI. The framing changes how you prompt — and what you get back.
Training-Time vs Context-Injected Knowledge: How LLMs Actually "Know" Things
Modern LLMs have two distinct sources of knowledge with very different reliability profiles. Understanding the split changes how you prompt — and how you spot hallucinations.
What NOT to Do With Modern LLMs
Certain interaction patterns reliably produce poor results — or worse, results that look plausible but are subtly wrong. Two anti-patterns that come from rushing.