Lab site · Built

Tested, with something you can run.

The demonstrations page is the library filtered to tested items with a live artifact. Same records, editorial presentation. Nothing here is signal or tried; if it is on this page, an experiment produced it and a measured result stands behind it.

repoTestedAgentic Memory System

Use a structured episodic store with summarised recall, not a raw vector memory

For any agent with more than ten sessions of history, store episodes with summaries and recall the summary first. Raw vector recall over transcripts loses on precision and latency above ~50k tokens.

TO Tom Okafor · 21 Aug 2026memory-adapter — one store, three harnesses

demoTestedAI Gateway

Route through a gateway you control; do not standardise on a vendor's

Every client pattern goes through a gateway the delivery team owns the config for. Vendor gateways are fine as a backend, never as the routing policy.

PR Priya Raman · 14 Aug 2026Routing table walkthrough: three gardens, six task types

evalTestedEval Harnesses

Every LLM judge ships with a human agreement score or does not ship

Do not report an eval number produced by an LLM judge unless the judge has a published agreement rate against a human panel on the same task. Below 0.8 Cohen's kappa the judge is not a measurement.

MT Mei Tanaka · 10 Aug 2026Judge calibration panel v2 — 300 items, three task types

repoTestedBackpressure

Put backpressure on agent fan-out before you put it on the model

Bound the number of in-flight sub-agents and tool calls at the orchestrator with a queue that applies backpressure. Model-side rate limits are the wrong place to discover you have a fan-out problem.

TO Tom Okafor · 25 Aug 2026bounded-fanout — orchestrator queue wrapper