The AI beat
AI engineering
Shipping agents and LLM systems that hold up outside the demo: grounding, evaluation, and the failure modes nobody screenshots.
When to abandon the low-code path
2026.04Why I rebuilt my procurement agent from scratch in week three of a four-month internship.
Anti-hallucination through tool grounding
2026.03A small experiment in forcing every claim from an LLM agent through a deterministic database query.
Notes on AI-assisted development
2026.02Patterns that have held up across four months of work with Claude Code.