Engineering Notes
The Ertir Blog.
Engineering notes from the people building Ertir: architecture, local inference, durable agents, and privacy you can check.
Posts
Letting a Local Model Search the Web Without Letting It Set the Bill
Ertir 1.3 adds crawler nodes: hosted Apify actors as tools a local model calls. We measured what a model sees, what it spends, and what breaks when it gets every field — and built the node around the answers.
Read the post · 4 min → coding agents ·Give Your Coding Agent Your Local Models: Ertir as an MCP Server
Ertir 1.2 ships an MCP server: Claude Code, Codex, and OpenCode can hand files to your local models without reading them, run and build your agents, and search your docs and code.
Read the post · 4 min → local inference ·Running Mixture-of-Experts Models Bigger Than Your Mac’s Memory
How Ertir’s MLX engine streams a mixture-of-experts model’s experts from the SSD, so a model that doesn’t fit in memory still runs with the same answers — and what it costs, measured.
Read the post · 5 min → architecture ·Building a Local-First Agent Platform Where the Vendor Is Never in the Middle
How Ertir keeps inference on hardware you control: two planes and one seam, logical model IDs, an engine manager instead of a model server, agents as content-addressed documents, and durable runs.
Read the post · 8 min →