The AI deflection rate that actually matters (and the three that do not)
Every AI vendor brags about a deflection rate. Most of those numbers are useless. Here is the one we publish, and why we publish it.
Stefan Goluwa · May 22, 2026 · 6 min read
Self-hosting an LLM in 2026: the cost math that made us do it
We run our own AI on our own hardware. Why we did it, what it cost, and at what scale it stops making sense.
Priya M. · May 12, 2026 · 11 min read
RAG vs fine-tuning for support AI: stop fine-tuning
The most common AI support pattern in 2026 is grounded RAG. Fine-tuning still has a niche. Most teams should pick one and not the other.
Priya M. · May 8, 2026 · 9 min read