Skip to main content
N/M experimental laboratories

Inspect experimental contracts where they run

Every lab connects a bounded language feature to runnable source, structured evidence, documentation, and an explicit trust boundary.

00

Evidence before claims

This directory separates runtime experiments, compiler inspection, and systems evidence. Each entry names the artifact it produces and the boundary that must remain attached when results are shared.

Connected surfaces
8 bounded laboratories
Evidence tracks
Runtime · Compiler · Systems
Execution model
Local-first, reproducible runs

01 / Laboratory index

Choose the lab by the question you need to answer

Start with the decision in front of you: compare executions, inspect compiler behavior, or review a systems artifact. Status labels describe product maturity, not scientific certainty.

01
Experimental 0.1

Runtime experiments

Experiment Studio

Run bounded algorithm presets across ideal, noisy, and mitigated lanes with explicit resource contracts.

Evidence produced
Versioned three-lane run record
Declared boundary
Finite-shot local simulation
Open lab
02
Preview

Runtime experiments

Seeded noise lab

Compare a fixed circuit under bounded local noise models and mitigation with deterministic seeds.

Evidence produced
Seeded distribution comparison
Declared boundary
Bounded local noise model
Open lab
03
Experimental

Runtime experiments

Scoped noise composer

Compose source-ordered gate, register, correlated-Pauli, and readout channels.

Evidence produced
Ordered channel contract
Declared boundary
Source-order local execution
Open lab
04
Experimental 0.1

Compiler inspection

Visual Oracle Builder

Compile bounded Boolean constraints into typed oracles with exact artifacts, resource evidence, and Playground handoff.

Evidence produced
Exact truth or sign artifact
Declared boundary
Bounded Boolean constraints
Open lab
05
Experimental 0.1

Compiler inspection

Circuit Analysis and Optimization

Inspect optimization diffs, target-native lowering, schedules, budgets, reversible lint, and measurement branches.

Evidence produced
Optimization diff and schedule
Declared boundary
Static lint, not a unitarity proof
Open lab
06
Experimental

Compiler inspection

Comptime inspector

Inspect compiler-folded dimensions, structural uses, source provenance, and execution limits.

Evidence produced
Fold and provenance report
Declared boundary
Compile-time constants only
Open lab
07
Experimental 0.1

Systems evidence

Backend Profiles and Calibration

Explore published target limits and compare two strict calibration snapshots without a live-hardware claim.

Evidence produced
Calibration comparison artifact
Declared boundary
Imported snapshots, not live devices
Open lab
08
Experimental

Systems evidence

QEC artifact lab

Inspect correction maps, syndrome tables, deterministic rounds, and verified feed-forward carriers.

Evidence produced
Correction map and syndrome table
Declared boundary
Bounded deterministic round trace
Open lab

02 / Evidence discipline

A laboratory result is a traceable experiment, not a hardware claim.

The N/M laboratory workflow keeps source, generated evidence, and the permitted interpretation together so a result can be reproduced and reviewed without overstating it.

  1. 01

    Fix the source

    Record the program, parameters, target profile, seed, and versioned contract before execution.

  2. 02

    Generate the artifact

    Produce a structured report, trace, comparison, or compiler artifact from the bounded local tool.

  3. 03

    Carry the boundary

    Publish the simulator, approximation, and verification limits beside every conclusion.