Chad Sakac

Another break from things quantum. Working on some stuff heads down listening to tunes that put me into a groove/flow zone. @Spotify throws in Blue Monday by @neworder dated, sure - but never old.

Is it just me, or does the @slack desktop client sometimes totally stop updating? The web client is way better, but man, sucks to have that all janky.

@BuffaloSabres @BuffaloSabres - truly brought a tear to this Canuck's eyes. Thank you.

That's how we do it in Buffalo. ❤️🤍

2) At the other end of the spectrum (world of business) this @McKinsey paper re the commercial reality of #Quantum. Looked at 300 (162 in detail) companies solving real problems with quantum tech right NOW. Their data matches my daily @IonQ experience. mckinsey.com/capabilities/m…

1) a pragmatically skeptical Scott Aaronson blog (congrats on NAS - that's a big deal!) on Q-Day here. Scott's view on this means a lot to me. He doesn't drink anyone's koolaid. "Will you heed my warnings NOW?" scottaaronson.blog/?p=9718

Two FASCINATING **different** things that highlight how fast things are moving the Quantum domain (feel lucky to leading that charge with my @IonQ teammates - disclosure).

If you are planning your trip for the QEC conference make sure to land on Saturday to attend our tutorial with Johannes Borregaard on Sunday. We will talk about fault-tolerant quantum computing with trapped ions.

@DesFrontierTech (And there are hints throughout the walking cat paper. An example is the effect on effective circuit execution time of high connectivity and walking cat states, along with massive parallel execution across cells).

@DesFrontierTech Most of all - none of this is the right way to think about it. What really matters is solving real customer problems, not academic theoretical arguments about point elements. We win that battle today. The walking cat paper shows how that evolves to winning tomorrow.

$IONQ In February, a customer in Singapore asked @sakacc about gate speed. They were using multiple quantum computing systems and wanted to understand why IonQ was outperforming the alternatives. Sakac is IonQ's Senior Vice President of Quantum Field Engineering. His answer was direct. Gate speed is the least important parameter. The metric that determines whether a quantum computer produces useful results is time to solution. Fidelity drives that metric, not gate speed. At some point the noise floor becomes a ceiling. The system is fast at producing results that are not useful. On a quantum Fourier transform algorithm, if you want a good answer, IonQ's system has a time to solution 1000x faster than the leading superconducting alternative. The superconducting system's time to solution swings by 100,000,000x depending on the accuracy required. IonQ's rises by one to two orders of magnitude. @sakacc/reality-rhetoric-customers-revenue-truth-3f793c836ec0" target="_blank" rel="nofollow noopener">medium.com/@sakacc/realit…

@nic_delfosse @DesFrontierTech Amazing work to the whole team, absolutely amazing. Celebrate (and then we’re all back to work - lots to do 😂)

If you had any doubt that we were taking fault-tolerant quantum computing seriously at IonQ check out our blueprint: 110 pages of compilation, error correction, micro-architecture, and a sufficiently fast decoder, validated with thorough simulations of all the components. Incredibly proud of the work of our team: Felix Tripier Woo Chang Chung, Jacob Young, Safwan Alam, Bryce Bjork, Aharon Brodutch, Finn Lasse Buessen, Nolan J. Coble, Thomas Dellaert, Dmitri Maslov, Martin Roetteler, Edwin tham, Mark Webster, Min Ye, John Gamble, Andrii Maksymov, John Paul Marceaux

@Cat_States Disclosure (IonQer). Thx for reading and glad you acknowledge credit where due (it’s a monster and the team has been working hard). Ack the observation. Rejoinder: do you think will anyone else be able to run a circuit with 10M-1B gates soon? (And we’re not stopping here)

IonQ just dropped a new architecture paper: the “walking cat” fault-tolerant design for trapped ions. Credit where it’s due: this is a real engineering blueprint, not a marketing deck. End-to-end spec with compiler, error-correction protocols and a real-time decoder. They’re using LDPC codes instead of surface codes, getting ~17× better qubit efficiency. 110 logical qubits from 2,514 physical qubits. Still a bit off the 12:1 ratio they need to hit next year, but it’s a start. The catch? One BIG one. The machine can only executive about 1 million operations per day. Their own headline benchmark - simulating a 100-site Heisenberg model, would take roughly one month of continuous runtime on a 10,000-qubit machine. One whole month. For a modest quantum physics problem. This isn’t cracking RSA anytime soon.

I think this could prove to be a big deal, an important milestone (time will tell). This (the @IonQ paper, not my blog which is my PoV on the topic) is worth pulling up a chair and reading: @sakacc/from-eniac-to-ionq-a-blueprint-moment-for-the-future-of-fault-tolerant-quantum-computing-2db2f7419d61" target="_blank" rel="nofollow noopener">medium.com/@sakacc/from-e…

@TechInnovationz Thanks for taking the time and effort to put this all down. I'm here for all those reasons and more. Working hard on all fronts 😀

@Adamisafakename @ionQ Oh, it's not random 😉

@sakacc @ionQ Do you think they just wrote some random stuff because they knew you'd be coming in? That's very common tbh

Came into the @IonQ Toronto office this am, and fascinating leftovers on the whiteboard from my engineering compatriots. I'm obviously not going to post a pic, but it's familiar territory to this "older gen" fella who is half electrical engineer, half comp sci. More on this soon.

@Russel1441065 @SPVLABS @Russel1441065 I'm always happy to talk to people. DM me (and please let me know more context) and we'll find a window

@sakacc @SPVLABS Maybe instead of wasting a day with this spv guy, give my guy 20 minutes. Much simpler to arrange... You won't change spv so let's move on to the positive! 20 minutes and I promise I will get verified so you know I'm not a bot lol.. just 20 minutes and ask him anything you want.

Ok, after this, I need to get back to my work, which matters more than twitpiss battles. Customers, revenue, results and published speak. More details in thread:

@Russel1441065 @SPVLABS 😉

@SPVLABS @sakacc