Technical Interview: QuEra

Brian N. Siegelwax


Cloud-Based Neutral Atom Quantum Computing

This is a companion article to Russ Fein’s article “Quantum Computing with Neutral Atoms,” which, in all honesty, is much better than this one. This article will be the first, however, in a series of articles reviewing the hands-on usage of publicly-available neutral atom products.

Thanks to Dr. Nate Gemelke of QuEra for spending more time with me answering my questions about the 256-qubit “Aquila” neutral atom quantum computer than I deserve. A follow up to this article will review my hands-on usage of this device, including illustrations where applicable.

The Interesting

I don’t normally have this section in my The Good, The Bad, The Ugly articles, but I’m not a hardware guy so I’m not going to classify everything I know.

Feynman-inspired. Aquila is directly inspired by Prof. Richard Feynman. It is a simulator meant to simulate systems.

Moving atom architecture. To use Aquila, you arrange the atoms to physically mirror the problem. You literally tell Aquila where to move individual atoms. I was really struck by how Dr. Gemelke kept saying “we move atoms around” so casually.

“Not hard to do.” The QuEra team includes only 40 people. The setup includes lasers and an optical table and, apparently, does not require a large workforce. From the sound of it, holding atoms would seem to be not as technologically challenging as we laypeople would assume.

Modesty. If neutral atom quantum computing really isn’t all that hard, which is the impression I got, where are everyone else’s devices? Believe me, I’m trying to access everyone else’s devices.

Illuminating. Aquila measures its neutral atoms by using 1,000 photons per atom to determine if it is there or not. Nothing touches the atoms besides light.

Ridiculously-fine control. QuEra engineers use Aquila to play atomic Tetris.

Topological qubits. Neutral atom positioning was used to create the first spin liquid (materials science). These spin liquids are probably the closest anyone has come to making topological qubits, which is famously of interest to StationQ at Microsoft.