Artiste-qb.net builds open-source tools that turn Python into a full quantum computing toolbox: quantum Bayesian networks, circuit compilers and simulators, quantum neural networks, and a curated cloud workstation for data scientists.
1997Quantum Bayesian nets first proposed by co-founder Robert Tucci
2017Inaugural CDL quantum machine learning cohort
17Public repositories on GitHub
270GitHub stars across the top ten repos
US 10,601,907Swarm compute patent, granted 2020
Background: a live quantum Bayesian network. Each node carries a complex amplitude; hue shows its phase.
Origin
A physics idea from 1997, rebuilt as a company.
In 1997 Robert R. Tucci posted a short paper to arXiv proposing Quantum Bayesian Nets: a way to draw quantum systems as the same directed graphs that Judea Pearl had made standard for classical probability.
Tucci built a Mac application called Quantum Fog to explore them, then paused it in 2006. A decade later, in September 2016, he and Henning Dekant, a quantum industry writer who had been covering the field on his blog Wavewatching since 2011, founded Artiste to bring the idea back as open-source Python.
The founding group grew to include Simon Bermudez and Tao Yin, with backgrounds spanning theoretical high-energy physics, machine learning, and biomedical engineering. The pitch was an "early bird" bet: build the developer tooling before quantum hardware went mainstream, so that companies could start experimenting on classical machines and switch to real qubits later.
Create and curate an open-source analytics platform that empowers data scientists to augment Python into a full-blown quantum computing toolbox.Company mission, via The Quantum Insider
Timeline
Three decades from paper to product.
Dates come from arXiv, the USPTO, the founders' own posts, GitHub commit history and press coverage.
“Quantum Bayesian Nets” on arXiv quant-ph/9706039
Robert R. Tucci generalizes Bayesian networks to the quantum regime.
US 5,787,236 and “How to Compile a QB Net”
Tucci patents a graphical computer method for analyzing quantum systems and describes compiling QB nets into sequences of elementary quantum operations, the seed of Qubiter.
Quantum Fog for Mac goes quiet
The original C++ Quantum Fog application stops development.
Artiste is founded
Henning Dekant announces the company with Bob Tucci, built around an open-source quantum Bayesian network framework on GitHub.
Signa.OS health collaboration
Work with medical researcher Andrew Deonarine on a health analytics platform using quantum-accelerated Bayesian networks for predictive and pandemic analytics.
Business relationship with D-Wave
Artiste starts developing software against D-Wave's API.
Creative Destruction Lab, first quantum cohort
Artiste joins CDL's inaugural quantum machine learning stream, kicked off at the Rotman School of Management in Toronto.
Swarm compute patent filed
Bermudez, Dekant and Horlacher file for a platform that distributes compute to web-capable devices.
Quantum Edward and BayesForge
A QNN library built on TensorFlow and Edward ships, and BayesForge goes live as a curated machine image on AWS Marketplace.
US 10,601,907 B2 granted
Patent protection runs until October 2037. The Quantum Insider profiles the company the same month.
Quantum in the browser
QFog-Pyodide and a JupyterLite Docker image move the tooling into WebAssembly. In April, The Quantum Insider names Artiste one of 12 quantum startups born and bred in Toronto.
Latest Qubiter update
Qubiter remains the organization's most-starred project.
New website under construction
artiste-qb.net shows a “coming soon” page with a Q3 2026 launch target.
Products
Four tools, one Python toolbox.
Everything runs on a classical computer today and targets gate-model quantum hardware from IBM, Google and others when it is ready.
Qubiter
Python · BSD-3 · ★ 123
Tools for reading, writing, compiling and simulating quantum circuits. Circuits live as human-readable text files with ASCII pictures you can edit by hand.
CSD compiler based on the Cosine-Sine Decomposition of linear algebra
Expands multi-controlled gates into CNOTs and single-qubit rotations
Model physical situations that show quantum behaviour using classical and quantum Bayesian networks. Built for quantum mechanics teaching and research in quantum AI, chemistry and cognition.
Graphical analysis of quantum measurement problems
Supervised learning with quantum neural networks. Train a layered QNN analytically on a classical computer with Black Box Variational Inference, then run the tuned circuit on real hardware, sidestepping noise during training.
Quantum frameworks IBM · Google · Microsoft · more
Artiste tools Qubiter · Quantum Fog
Data science Python · Jupyter · analytics libraries
Linux machine image Amazon Web Services
A Linux machine image that curates the best open-source software for data scientists and for quantum computing and computational mathematics practitioners. Launch it and the major QC frameworks are already installed.
One-click cloud workstation on AWS
Free tier path with Python and Qubiter preinstalled
From first programs to quantum-ready corporate systems
A three-qubit state-vector simulator running in this page. Click a slot to cycle through gates. A ● control in the same column turns every other gate there into a controlled gate, the multi-controlled style Qubiter compiles down to CNOTs.
Circuit
Cycle: empty → H → X → Z → T → ● control. Time runs left to right.
State vector |q0 q1 q2⟩
phase 02π
Intellectual property
Turning every browser into a compute node.
US 10,601,907 B2 describes a platform that splits a heavy workload into subtasks, farms them out to a swarm of ordinary web devices running GPU-accelerated WebGL, and reassembles the answer, with a blockchain ledger recording who computed what.
subtask outresult backledger block
Title
System and method for platform to securely distribute compute workload to web capable devices
Inventors
Simon Adrian Bermudez Hernandez, Henning Dekant, Scott Horlacher
Assignee
Artiste QB Net Inc.
Filed
2017-09-22
Granted
2020-03-24
Expires
2037-10-26 · Active
Prior art
BOINC, Project Bayanihan, CrowdCL, plus WebGL, WebAssembly and ECMAScript specs
Split
The server receives a workload request and cuts it into subtasks.
Swarm
Phones, tablets, consoles, laptops and PCs compute in WebGL-enabled browsers.
Secure
Digital signatures, encryption, shared secrets, and smart chargers as hardware dongles.
Ledger
Smart-contract blockchains keep an immutable audit of every executed task.
Scale
Microservices spawn requestor and swarm instances as demand rises.
Open source
Built in public on GitHub.
Seventeen public repositories. Stars and forks for the ten most visible, as of September 2026.
Stars and forks
starsforks
qubiter · the most active codebase, last pushed December 2023. Mixed licensing: BSD-3 with patent rights, GPLv2 for the CSD compiler.
Python-CS-Decomposition · Fortran bindings that make the Cosine-Sine Decomposition available to Python, powering Qubiter's compiler.
QFog-Pyodide and jupyterlite-docker · 2021 experiments running the whole stack in WebAssembly, with no server needed.
Entanglement-Lab · software for calculating entanglement of pure and mixed states.
os_quantum_software · a fork of the community's curated list of open-source quantum software.
Ecosystem
One of twelve quantum startups born in Toronto.
In 2021 The Quantum Insider mapped Toronto's quantum scene, much of it grown through the Creative Destruction Lab. Select a company to see its focus.
Databases disagree on the founding date: Tracxn lists 2015, Crunchbase lists November 2018, and the founders' own blog announces the company in September 2016. This page follows the founders' account.