| Aug | The group opens at NUS in 2027 and I am hiring: two postdoctoral researchers and PhD students, in Computer Science and in Physics. If verifiable quantum advantage, quantum many-body physics, or machine learning for quantum computers is your thing, see joining the group for deadlines and how to apply. |
| Aug | Heuristic Quantum Advantage with Peaked Circuits was named #1 Best Paper of the 56 in the Quantum Algorithms track at IEEE Quantum Week (QCE) 2026. A 2000-gate instance that Quantinuum’s H2 solves in under two hours extrapolates to years on Frontier — and unlike sampling claims, the answer is checkable. With the team at BlueQubit. |
| Jun | How do you tell which mixed-state phase a noisy quantum computer is actually in? With Timothy Hsieh and Yijian Zou at Perimeter and Yong Baek Kim at Toronto, we answered it on Quantinuum’s H1 with Rényi correlators and a variational decoder — now out in Nature Communications. |
| Jun | Delighted to share that I will join the National University of Singapore as an Assistant Professor in December 2026, jointly appointed in Computer Science and Physics, where I will start a group on quantum information, quantum many-body physics, and AI. |
| May | Can a quantum computer learn how to compile its own dynamics? Scalable quantum dynamics compilation via quantum machine learning is out in Physical Review Research — with Roeland Wiersema (Vector & Waterloo), Juan Carrasquilla (ETH Zürich), Lukasz Cincio (Los Alamos) and Yong Baek Kim (Toronto). |
| Mar | Neural quantum states that respect biorthogonality can take on two-dimensional non-Hermitian systems: Biorthogonal Neural Network Approach to Two-Dimensional Non-Hermitian Systems in Physical Review Letters, led by Massimo Solinas with Brandon Barton and Jannes Nys in Juan Carrasquilla’s group at ETH Zürich. |
| Mar | Lectured on quantum complexity and quantum advantage at the School for Near-Term Quantum Computing in Benasque. |
| Dec | Evaluating Large Language Models in Scientific Discovery (arXiv:2512.15567) — a large multi-institution effort led by Zhangde Song, asking whether LLMs can actually run the loop of hypothesis, observation, and revision, rather than recite decontextualized facts. |
| Dec | Moved to Europe: started as a postdoctoral fellow with Dmitry Abanin, superposed between EPFL and Princeton, and joined BlueQubit part-time as a research scientist. |
| Nov | Took part in a program at the Kavli Institute for Theoretical Physics in Santa Barbara. |
| Oct | Two preprints on peaked circuits — near-term feasible, classically hard, and verifiable, which few quantum-advantage proposals manage all at once: Complexity and hardness of random peaked circuits (arXiv:2510.00132, single-author) and Heuristic Quantum Advantage with Peaked Circuits (arXiv:2510.25838) with the BlueQubit team. |
| Jul | New preprint: Circuit compression for 2D quantum dynamics (arXiv:2507.01883) — led by Matteo D’Anna, with Roeland Wiersema, Manuel Rudolph (EPFL) and Juan Carrasquilla (ETH Zürich). |
| Apr | We watched a supersonic mode travel across an 18-site fermionic chain on Quantinuum’s H1 — a correlation front that the Lieb–Robinson bound forbids in any Hermitian system. Observation of a non-Hermitian supersonic mode on a trapped-ion quantum computer, in Nature Communications with Juan Carrasquilla and Yong Baek Kim. |
| Mar | Mixed-state phase transitions in spin-Holstein models in Physical Review B — with Brett Min and Dvira Segal at Toronto, and Yuxuan Guo and Yuto Ashida at the University of Tokyo. |
| Feb | It is not just how much magic a circuit has, but where you put it: pile every T gate into a single layer and the circuit becomes classically simulable. Classical Simulability of Quantum Circuits with Shallow Magic Depth, with Yifan Zhang (Princeton), appeared in PRX Quantum and was selected for the journal’s International Year of Quantum (IYQ) Collection. |