My name is Bogdan Reznychenko, I am a Quantum Software Engineer at Photonic Inc based in Greater Vancouver area in BC, Canada.
This repository was meant to show my progress in coding, life and the rest of 42. The projects here are:
- QGhostbusters game developed with Lucas Froguel for Quantum Games Hackaton 2023.
- foobar: my solutions to the medium and hard problems of the Google foobar challenge. Basically, these are LeetCode style problems.
- quantum_control: a personal project,
where I use reinforcement learning to put and keep a qubit in a state I want it to be. I also had some fun making
ipywidgets
interface to show the result. - In quantum_computing I experimented with
qiskit
and quantum machine learning by solving a simple IBM quantum challenge problem. - pennylane fork was created to fix an issue by merging this pull request. You can look at it to see my work style (lots of commits for a simple problem, but I write docs and unit tests mostly without reminders).
- there is a qutip fork, I contributed some precision and performance improvements for transformations between Kraus and Choi representation of quantum channels.
- divinum-officium fork is a slow work in progress. The original repo is a multilingual translation of Roman Catholic Mass and Office, and I forked it to add my translation into Ukrainian (that were originally in LaTeX and for a non-Roman Western rite, so some adapting is needed...).
I also have a GitLab account, where I participate in another Divine Office project, but the repo isn't public. We basically make templates to make those pdfs with as much automation as possible.
You have probably been there, but if not, go see my LinkedIn page. But in short, I have worked on different projects on:
- Currently I work as a Quantum Software developer at Photonic Inc, where I own a tool for various simulations of physics a quantum system used by the company as a quantum computer building block.
- Also, I developed a bunch of algorithms for simplifying work with wide and deep quantum circuits for the company.
- On my previous job I led a CS part of an international interdisciplinary research project on quantum-assisted computation. I developed architecture and a compiler for a quantum co-processor and evaluated potential computational benefits of the device.
- In my PhD research:
- I developed theoretical description of quantum measurement back-action and used it to control qubit state (in theory).
- Collaborated with different experimental teams in description of experiments with qubits (QDs and transmons), including characterization.
- Simulated experiments in Python and Matlab.
ML @ Probayes:
- Time series forecasting with various underlying models (neural networks, Bayesian and Hidden Markov models, more standard tools like gradient boosting, linear regression etc). The most interesting results so far: the week I joined a project that has been going on for half a year, I suggested a different approach, and it raised the R2 score from about 0.01 to 0.5.
- Causal inference: establishing an effect of a particular factor (way of contact a client, prices of products) on a desired outcome. In all the projects I identified and implemented the most appropriate models from research papers into shipping products.
- Reinforcement learning: drone flying in a simulation. So far the agent is doing what I want it to do. Also there is quantum_control personal project.
You can also check out my publications at Google Scholar. They are mostly in Physics, plus one on the interface between CS and Quantum Physics.
You can write to me at rezniuk at gmail.com, or @theodotk in Telegram, or at LinkedIn.