06 May D-Wave Making quantum computing accessible
Originally published by Invoke Digital
Working with quantum computing pioneer D-Wave to develop a user-centric data visualization tool that helps engineers understand and benefit from D-Wave’s advanced technology.
At the cutting edge of their field
D-Wave is a pioneering quantum computing company based here in BC. Like many of our clients, they’re at the forefront of their field — with their quantum technology securing over $200 million in funding and their customers including the likes of Volkswagen, DENSO, and national labs including Oak Ridge National Laboratory and Los Alamos National Laboratory.
While their technology is incredibly advanced, they sought Invoke’s product expertise — our ability to uncover and analyse opportunities to create experiences at the intersection of desirability, viability, and feasibility — to translate this world class technology into a world class education experience for the engineers that use it.
The core of our engagement was working on D-Wave’s inspector tool and the visualization framework within it that helps customers understand their quantum problems. When we began working with them, the tool was functional on the small scale — visualizing the Chimera Graph (2000 qubits with 6000 couplers connecting the qubits) — but wasn’t particularly user friendly. A large-scale version was underway to visualize the Pegasus Graph (over 5000 qubits and 40,000 couplers), but it wasn’t yet accomplishing what the D-Wave team was targeting.
The scale and complexity of visualizing the Pegasus Graph alongside other important features would prove to be daunting — a challenge our team was up for.
Uncovering opportunities through discovery
To push their product to the next level, we collaborated closely with D-Wave and their customers, facilitating activities to help us uncover opportunities in an industry that was new to our team.
We began with understanding the fundamentals of quantum computing, the business objectives of D-Wave, and the background of their inspector. To do this, we conducted onboarding sessions complete with quantum theory and tool demos and gained insights through internal stakeholder interviews and reviews of comparable products. It is equally important to consider the user by ensuring our solution would meet their needs. Accomplishing this meant understanding their motivations and mapping their experience with D-Wave, done using empathy maps and journey maps.
Validating a product direction
With a lot of inspirational ideas and opportunities emerging from discovery, we developed product hypotheses that helped us prioritize features for an MVP offering of the inspector.
A simple and easy to use interface was important, but it couldn’t sacrifice functionality. It still had to provide many functions (such as graph views, error console, tips, workspace views, panels, detail specs, and more) and include the right visualizations (ways to visually represent different states of qubits and couplers) for the four graphs that make up the inspector. All of this had to happen while supporting multiple view options that allow people to inspect their submitted problem and solution in different ways.
Equally important to how we determine the features that would address user needs is determining how and when those features are implemented. Our tech team created detailed tech scoping of all proposed features, recognizing that there was an opportunity to streamline what would be the MVP solution. In collaboration with D-Wave, we utilized the MoSCoW method to prioritize feature sets for release and a product roadmap. Going through this process not only helped us build the right features, it ensured we could quickly decide which additional “could have” features to implement when we had time at the end of the project.
For D-Wave, we produced sketches and clickable wireframe prototypes to showcase early iterations of how they would use the inspector interface and what information they’d expect to see in the visualizations.
Over the course of this project, we tested and interviewed with 30 new and existing customers across the globe. From showcasing early iterations and ideas with sketches and clickable wireframe prototypes, to the many design iterations of the visualization graphs, to usability testing on high-fidelity UI, we made sure to conduct testing at least once a month. This helped ensure customers were being listened to and having their feedback incorporated. Members of the D-Wave team often joined these testing sessions; this gave them first-hand learnings and greater confidence in their product decisions.
A collaborative launch
At Invoke, we frequently work in conjunction with our client’s internal tech teams. D-Wave was no different, as we collaborated with their systems and software, tools, QA, brand and marketing, and Leap management teams. As we absorbed their quantum expertise, they smoothly adopted our user-centered process — we even helped them with their very first design hire to ensure that process and practice continues in their organization.
Working with internal tech teams means finding ways to ensure new solutions mesh seamlessly with those that are existing. So, instead of delivering all work in a single codebase, we delivered a flexible collection of libraries and components that can be reused within each other (e.g., the Graphing Library, Inspector Components, and Demo Application). Additionally, there are embeddable components that can be utilized in multiple contexts, meaning the problem inspector can be used across the D-Wave product suite.
To prepare for the inclusion of the visualizer and inspector in the launch of Leap 2 and their integrated development environment, we’ve continued to work closely with D-Wave on refining the details, functional QA, and finding ways to improve the experience.
At the outset of the project, visualizing our next generation quantum architecture was such a challenge we weren’t confident it could be done. That was the immensity of the task we were facing. Invoke’s innovative approach helped us navigate the complexity for what is now the state-of-the-art problem inspector in the quantum computing industry.