A speculative biometric safety tool, designed as a "LinkedIn for past crimes" - giving people secure access to verified criminal records so they can make informed decisions about who they meet, work with, or trust.
At a glance
- Role: Designer (team of 4)
- Methodology: Lean UX, value-driven design
- Tools: Figma
- Presented at: Communitech, Waterloo, Ontario
I owned the onboarding and ID-verification flow for organizations, and led the depth pass on the criminal records profile — the two screens doing the heaviest lifting in the product. I also led our team's ethical evaluation through cultural relativism, which shaped how the tool adapts across cultures and life stages.
Why Sora
The team kept circling the same uncomfortable observation: people - especially women and vulnerable individuals - make daily decisions about who to trust with almost no reliable information. We wanted to design something that could put that information into their hands responsibly, without it becoming a surveillance tool.
Sora - Yoruba for "to be cautious" - became the framing. A biometric-enabled app that connects users to verified criminal records through global databases.
Users can access their own records, share verified versions with people they trust - a friend, a date, a new colleague - and pursue restorative-justice options to clear or amend past entries. Organizations can vet individuals through sensor-based scan, with consent controls baked in.
The hook we kept telling ourselves: a LinkedIn for past crimes.
Approach: ethics before pixels
Most apps I'd designed started with user flows. This one started with ethical theory.
We worked through cultural relativism, Kantian ethics, and Friedman's envisioning cards to interrogate the product's value across stakeholders - direct (users, organizations), indirect (people whose records might appear), and cultural (criminal record means different things in different societies). Speculative exercises — Tarot Cards of Tech, and imagining a "Black Mirror" episode set inside Sora - forced us to see the dark paths the product could take if we got it wrong.
Then three Lean UX cycles: Think, Make, Check, refining against feedback each turn.
A constraint worth naming: the sensitive subject meant we couldn't recruit real users with criminal histories. We leaned on secondary research and ethical scrutiny, but it's a real limitation - and one I'd want to address if I returned to the project.
A system, not just an app
Sora is two things: an app that people carry, and a small physical sensor that organizations mount at entry points. Splitting the product across software and hardware wasn't cosmetic - it was structural, because Sora has to serve two very different consent situations.
Individual to individual. When a person wants to share their record with someone they trust - a friend before meeting up, a match from a dating app, a new roommate - the app handles it entirely. The record owner grants access from their own device; the recipient views it on theirs. Consent is deliberate, digital, and reversible.
Organization to individual. When an organization needs to verify someone at an entry point - an event, a venue, a workplace - the app alone isn't enough. The scanned person and the organization need a shared physical moment where consent can happen in the open, in front of both parties. That's what the sensor is for. The person places their finger on the sensor; that touch is the consent. Once verified, their profile appears on the organization's own device, not on the sensor itself. No touch, no scan. No hidden capture, no ambient surveillance.
The two flows share one underlying rule: nobody's record gets seen without an explicit, embodied act of consent from the person it belongs to. Digital consent for peer-to-peer sharing, physical consent for organizational vetting.
The sensor is deliberately small and discreet - roughly the size of an access-control reader you'd find at any office door. A larger scanner would have read as surveillance apparatus, which is the opposite of what we wanted. Small also meant deployable at scale: mountable on a turnstile, a podium, a wall panel, or a will-call desk without significant retrofit. And smallness constrained capability by design - the device can read biometric input and communicate with the app, but it can't record video, capture audio, or gather ambient context. The form factor is a hardware-level privacy commitment.
The split also created a clean division of trust: the sensor belongs to the organization performing the check; the app belongs to the person being checked. Two different users, two different surfaces, two different sets of consent controls.
What I designed
Organization onboarding & ID verification. A flow that lets vetted organizations register, prove legitimacy, and request scans - with friction calibrated to discourage misuse without blocking legitimate use cases. Government ID scanning replaced manual entry to cut errors and speed onboarding.
Criminal records profile. The most sensitive screen in the product. I worked on what to show, what to summarize, how to handle a clean record versus a complex one, and how to make a serious surface feel humane rather than punitive. Information density was the central design problem.
I also contributed to the design system rollout in the third sprint, applying color and typography decisions across screens for consistency.
Where it landed
We presented Sora at Communitech in Waterloo - a sharper accountability moment than a classroom critique. The feedback shaped priorities for future refinement, especially around personalization of profile views and tighter access controls on the organization-scan feature.
Reflections
Ethics is a design tool, not a checkbox. Envisioning cards and speculative exercises early on changed structural decisions, not just surface ones. I want to bring this into every project where the stakes are non-trivial.
Sensitive products demand more user input, not less. The work we couldn't do with real stakeholders is the work I'd most want to do next. Designing safety tools without lived experience in the room is a constraint I'd resist on the next project of this kind.
Some surfaces carry more weight than others. A few screens were doing most of the ethical work. Recognizing those early and protecting their design time mattered more than evenly distributing effort across the whole app.