Trevor Vincent, Phd Email & Phone Number

Helping PsiQuantum develop high-performance quantum application software.

Toronto, Ontario, Canada

Between Nov 2022 and Nov 2023 I developed software for QC Design, a company specializing in quantum error correction. A few highlights:- Developed a high performance union-find decoder for QC Design (https://qc.design). - Developed a fast C++ graph library for QC Design. - Developed a Kokkos-accelerated Clifford state simulator for QC Design.- Sped up a core QC Design library by 300x

Toronto, Canada Area

- At Xanadu, I was the lead High Performance Computing (HPC) engineer on the Borealis Quantum Advantage demonstration (Nature 606 (7912), 75-81) which made the front page of the Globe and Mail (Canada's most widely read newspaper). Borealis is the world's first quantum computer on the cloud with a peer-reviewed advantage claim. - I was the inventor (patent pending) and lead developer of the task-based tensor network simulator Jet (github.com/XanaduAI/jet). The details can be found in it's paper published in the Quantum journal.- Jet has been used in a tensor network / quantum machine learning paper published in Nature Scientific Reports by a research group at BMW. - I later lead development of the first AMD-GPU quantum circuit simulator Pennylane-LightningKokkos (github.com/PennylaneAI/pennylane-lightning-kokkos) and created the first code performance regression analyzer call Benchmark, which currently runs at benchmarks.pennylane.ai- I also managed one intern (Isidor Schoch) through a successful project involving the use of the Kokkos library for high-performance quantum statevector simulation.

Funded by NSERC CGSD and NSERC CGSM scholarships. I developed next-generation numerical algorithms for solving the Einstein field equations and tested them out in a 100,000 line code. I also worked on a team of 10+ developers to construct a framework called SpECTRE for next-generation numerical relativity (solving the Einstein field equations on supercomputers). On top of this I worked for the Nobel-prize winning collaboration LIGO, simulating the collision of neutron stars.

With funding from a CIHR undergraduate research award, I developed C code for a Bruker magnetic resonance imaging machine to apply oscillating magnetic fields (the machine could not do this through the default software). On top of this, I ran for the first time, ultra-low magnetic-field diffusion simulations with my CUDA C++ code.

With funding from a NSERC undergraduate research award, I developed CUDA (GPU) accelerated diffusion-magnetic-resonance-imaging simulations and a Markov Chain Monte Carlo code for extracting axon properties in brain matter from magnetic resonance signals. This work was later published in a peer-reviewed journal.

With funding from a NSERC undergraduate research award, I developed diffusion-MRI Monte Carlo simulations. These simulations involve the diffusion of water molecules in brain matter coupled with a solver of the Bloch equations to simulate the effects of magnetic resonance imaging. This work was published in a peer-reviewed article (see trevorvincent.com and my resume for details).

Winnipeg, Canada Area

With funding from a NSERC undergraduate research award, I developed segmentation algorithms for magnetic resonance images of mice brains. These algorithms were used to determine if Alzheimer gene-models induced negative volumetric changes in the brain. This work was published in two different peer-reviewed articles (see trevorvincent.com and my resume for details).

Doctor Of Philosophy - Phd, Physics

PhD work was funded by an NSERC CGSD scholarship ($105,000) and a NSERC CGSM scholarship ($30,000).

Education record

Bachelor Of Science - Bs

Funded by 15 scholarships (incl. 3 NSERC USRA, 1 CIHR). Winner of the Governor General's Medal for highest graduating GPA in the university.