Can we still explain the experiment two months later?
A result is easier to revisit when its question, code and conditions stay together. That simple idea shapes the experiment record in QFlow Studio.
Two months later, you remember the experiment better than its files do. You can still explain why the question mattered. Finding the circuit that produced the chart is harder. The result is in a dashboard, the code is in a notebook, and the conversation that joined them has slipped down a message thread. Returning to the research begins with searching for your own work.
Careful people can end up here too. Each tool kept its own piece; none kept the relationships. In QFlow Studio, I am trying to make those relationships part of the ordinary workflow, so remembering everything is not another job for the researcher.
Keep the question beside the result
The record follows the brief, circuit, source, route, run and evidence. In ordinary language: what we wanted to find out, what we built, what code ran, where it ran and what came back. The question gives the result meaning; the conditions tell us how far that meaning reaches.
Source stays visible across Qiskit, Cirq and OpenQASM. Provider choice and calibration context stay with the run. That makes a later comparison more than a search for similar-looking screenshots.
Sharing a result should be straightforward
A collaborator should be able to inspect the result without receiving provider credentials. Keeping secrets separate from shareable experiment material makes that easier from the start. It is better than relying on someone to hide a console field every time they take a screenshot.
A readable record also helps a reviewer ask better questions. They can follow a choice back to the original brief, or see which conditions changed between runs. The benefit is a shorter route to the conversation, rather than more paperwork.
AI can help prepare; people choose to run
AI may help draft an experiment. Sending a job to hardware still needs user approval, because it spends money, takes a queue slot and creates a real run. Keeping that choice explicit helps the researcher know what has been prepared and what has actually happened.
QFlow builds around existing quantum tools. It does not make their calculations more powerful or establish a quantum performance gain. The aim is more everyday: when someone returns to the work, they should be able to understand it and carry on.
- The original question
- The code and run conditions
- The result we can revisit
Sources
- QFlow Studio: Quantum Workflow and Learning Studio qflow.studio