Dec 15th 2025.
​​
It has been two years already and what a ride!
​
IonQ grew from 200 employees to more than 1,000 and acquired several startups. I am particularly excited about the acquisition of Lightsynq, which brings the potential for faster interconnects (crucial for scaling), and Oxford Ionics which gives us two-qubit gates with fidelity > 99.99, a major experiment breakthrough that opens the way to all sorts of new QEC schemes that do not work with more noisy qubits.
One of my main motivations for joining was the opportunity to build the QEC team. I was lucky to have very strong researchers and engineers transferring internally from other teams – thank you Edwin, Aharon and Felix for trusting me. External hires of exceptional researchers like Min and Nolan reinforced the team and bring fresh ideas. Overall, John Gamble provided us with an ideal work environment within the architecture team that he is leading. We have fantastic collaborators and a very light admin burden which allows us to move fast. Today, we are a team of six with number seven starting soon and many more to come.
​
Our internship program is now up and running (thank you Annie!). Greg was the first QEC intern and worked on partial error correction [3]. Next, Arda, Evan and Nolan joined the team and worked on cyclic HGP codes [5], distributed quantum computing [10] and QEC for chain losses [7]. I am looking forward to welcoming our third generation of interns next summer. Our interns are typically remote for a 12-week project, and we make sure to design an impactful project that is doable within this short timeline. This generally involves a mixture of QEC design and simulations. Feel free to reach out if you are looking for an internship and you have solid QEC experience.
Over the past two years, we made progress in several directions:
-
Partial QEC: We designed a new approach to partial quantum error correction called CliNR [1], we optimized it and ran a CliNR experiment on a 36-qubit machine [2] and we introduced a recursive CliNR scheme [3].
-
New codes: We designed new codes with BB5 codes [4] and cyclic HGP codes [5].
-
Code layout: We designed efficient QEC circuits for a long chain of trapped ions [4] and we proposed a modular design that applies to long chains or short chains [6].
-
Chain losses: We introduced a protocol to correct chain losses [7].
-
Stacked quantum memories: We also explored long-term ideas like with our stacked quantum memory project [8].
-
New decoder: We proposed a new decoder that fits the needs of a trapped ions and that can be implemented with off-the-shelve CPUs without any specialized hardware [9].
-
Distributed quantum computing: Finally, we demonstrated a potential advantage in distributed quantum computing with slow interconnects which could be practically relevant when we start connecting multiple QPUs [10].
​
Quantum error correction used to be a mostly theoretical field which I enjoyed very much as a mathematician. There is nothing more satisfying than a clean proof of a non-trivial result based on intuitive definitions and lemmas. However, over the past few years, we have seen incredible experimental progress from the whole community. Now, thanks to Oxford Ionics hardware (and their amazing team), we have a real shot at building an impressive fault-tolerant machine and I won’t let this opportunity go to waste.
Feel free to reach out if you are excited about the idea. Our team is growing. Exciting time to work in QEC at IonQ!
​
References:
[1] https://arxiv.org/abs/2407.06583
[2] https://arxiv.org/abs/2504.13356
[3] https://arxiv.org/abs/2511.22624
[4] https://arxiv.org/abs/2503.22071
[5] https://arxiv.org/abs/2511.09683
[6] https://arxiv.org/abs/2508.01879
[7] https://arxiv.org/abs/2511.16632
[8] https://arxiv.org/abs/2411.09173
[9] https://arxiv.org/abs/2512.07057
[10] https://arxiv.org/abs/2512.10693
​
​​
See what I wrote when joining IonQ.
​