Skip to main content
iD01t Productions · Longueuil, Québec · ORCID 0009-0000-3465-3753

About QECTOR

Independent quantum error correction research, engineering, and commercialization, built by Guillaume Lessard with full transparency, reproducible artifacts, and no speculative claims.

Creator

Guillaume Lessard, software engineer and researcher based in Montreal / Longueuil, Québec, Canada.

Operating as the QEC software division of iD01t Productions (founded 2012), Guillaume focuses on quantum error correction software engineering, building high-throughput decoders and advanced validation tooling.

Publishing aliases:Guillaume Lessard(engineering)El'Nox Rah(research / authorship)DJ iD01t(music)

Research Background

  • CSS stabilizer codes: surface codes, LDPC, qLDPC
  • MWPM (Blossom), Belief-Matching, BP-OSD decoding
  • Custom CSS code topology research and GF(2) eligibility analysis
  • Bivariate bicycle qLDPC codes
  • Reproducible provenance and SHA-256 sealed artifacts on GitHub
  • Monte Carlo threshold analysis and reproducible validation workflows

Software Stack

  • Rust / PyO3: core decoder engine
  • Python 3.9-3.14: API, CLI, benchmarking, simulation
  • Stim: circuit-level noise simulation
  • OpenCL / CUDA: GPU-accelerated batch decoding
  • PyPI distribution: qector-decoder-v3
  • Ed25519 offline cryptographic signature verification
  • CustomTkinter / FastAPI: QECTOR Workbench desktop interface & REST engine

Engineering Philosophy

  • Reproducibility first: public claims link to supporting artifacts where published
  • Honest documentation: Decoder is Source-Available (not free); Workbench GUI app is free
  • Simulation-validated before any public claim
  • Full provenance trail: SHA-256 bundles
  • No vendor lock-in: pluggable CodeProvider architecture
  • Transparent validation reports with non-pass classification

QECTOR Product Scope

This site and QECTOR Decoder v3 + Workbench constitute the high-performance decoder library and professional GUI.

  • QECTOR Decoder v3: Rust/Python multi-algorithm QEC decoder library (Source-Available)
  • Free QECTOR Workbench: live Windows and Linux releases with 85-tool MCP servers and 19 named decoder kinds (17 concrete configurations plus AutoDecoder and Auto Router).
  • Documentation & Education: Public methodology, validation guidance, and "Mastering QEC" book context (see the decoder repository for primary evidence).

Project Timeline

2026-06-23

The public qectorlab.github.io repository was created.

2026-06-24

The qector-decoder repository was created and v0.5.0 / v0.5.1 were first recorded on PyPI.

2026-06-24 to 2026-07-26

The verified PyPI release train progressed from v0.5.0 through v0.6.9. Dates are listed on the Changelog page from PyPI upload metadata.

2026-08-02

QECTOR Decoder v3 v0.7.0 was published on PyPI.

2026-08-06

QECTOR Decoder v3 v1.0.0 reached its first stable release, with the official reference manual (DOI 10.5281/zenodo.21941046).

2026-08-11 to 2026-09-04

QECTOR Workbench Windows v1.0.0 and v1.0.1 were followed by verified v1.0.7 releases for Windows and Linux (2026-09-04).

2026-08-15 to 2026-08-24

The Claude Plugin repository was created on 2026-08-15; plugin v1.0.6 was published on 2026-08-24.

2026-11-06

QECTOR Decoder v3 v1.0.1 was published with 18 wheels for Python 3.9-3.14 and signed build provenance, alongside Claude Plugin v2.0.1 and the new commercial plans.

Transparency Statement

QECTOR is an independent R&D project, not backed by a quantum hardware company, VC funding, or a university lab. No hardware-specific measurement figures are published on this site, because results depend on your hardware. The decoder reached its first stable release on 2026-08-06 (v1.0.0); the current release is v1.0.1. QECTOR Decoder is source-available under the published license. Workbench release terms are platform-specific. We publish validation guidance and known limitations openly. Documented behavior is what ships.