What Breaks When You Self-Host an LLM
Four handoffs, nine metrics, and one script that shows you which one is already broken
The AI Engineer
Become dangerously good at AI Engineering.
Live Last read · last published · next check
Latest posts
What is a Reranker?
A reranker is a second scoring pass over results a search has already returned.
What is an Eval?
Why it looked better is not evidence
Why Your Multi-GPU Training Job Keeps Dying
Four ways a small multi-GPU run kills itself, and the logging that tells you which one it was
What is Inference?
Why reading is fast and writing is slow
Superpowers Vs GSD Vs Compound Engineering
Three agent frameworks that disagree about exactly one thing
What is Chunking?
You set your RAG quality ceiling at ingest, before the first query runs
RAG vs Fine-Tuning
The cheap fix works more often than you'd guess (mostly
Should You Self-Host Inference?
The honest economics of running your own models, and the breakeven most teams get wrong.
AI Coding Tools: What Changed in the Last 6 Months
A 10-minute update on what shipped, what broke, and what it costs now
Mem0 vs Zep vs Letta vs cognee
Four architectures for agent memory, and how to choose the one that fits your stack