Test everything twice.1
Axiom Lab runs open, replicable research in materials, energy storage, and measurement — and publishes the failures alongside the wins.2
1. then once more, to be sure · 2. misses included
Explore the programs Our methodResearch programs
The periodic table of what we're up to.
Each program gets a tile: symbol, headcount as its number, and funding runway as its mass.
Materials
Self-healing polymer coatings and where they actually break.
runway 2.4 yr · 3 preprints 9 EsEnergy storage
Sodium-ion cells for grid duty: cycle life over energy density.
runway 3.1 yr · 5 preprints 6 MeMetrology
Cheaper reference instruments so replication stops being a luxury.
runway 1.8 yr · 2 preprints 4 RpReplication unit
We re-run other labs' headline results. Batting average: published.
runway 2.0 yr · 11 reportsFindings
Recent output, wins and misses alike.
Sodium-ion pouch cells survive 4,000 cycles at 92% retention
Full dataset, cell teardowns, and the two batches that didn't make it.
published AX-2026-011Replication: the "room-temperature" coating claim doesn't hold below 40% humidity
Original effect confirmed in a narrower envelope than reported.
published AX-2026-017A $220 reference thermometer with metrology-grade drift
Build guide, calibration protocol, and 90-day drift data. Preprint open for review.
preprintMethod
Pre-registered. Open data. Failures included.
Every study is registered before the first measurement, every dataset ships with the paper, and every null result gets the same typesetting as a breakthrough. Science is the log, not the highlight reel.
$ axiom publish --include-nulls --always