LongBrain Blog

Writing from the team building persistent memory for AI.

Memory February 2026 5 min read

Why AI Assistants Keep Forgetting You

Most assistants can reason well in a single session but lose context across days and weeks. We break down where memory collapses in real workflows, and why retrieval quality matters more than raw model size when users expect continuity.

Engineering January 2026 8 min read

Building in Public: How We Made a Memory Layer for AI

LongBrain started as a simple idea: persistent context should be reliable, observable, and easy to integrate. This post covers the product decisions, infra tradeoffs, and public iteration loop that shaped our memory layer.