Research and development at the intersection of user experience, security and privacy, and decentralized infrastructure. Goal: publish reproducible UX research, prototypes, and tools that shorten time-to-value in healthcare, cybersecurity, DePIN, and accessible AI.
- Patient intake, surgical coordination, clinician decision support under cognitive load
- Accessibility in clinical UIs (WCAG 2.2 AA), audit trails, safety-critical interaction
- Data handling without PHI leakage; de-identification and synthetic data workflows
- Usable cryptography: key management, MFA, threshold shares and recovery
- Interfaces for consent, auditability, data minimization, and policy enforcement
- Operator tooling to reduce misconfiguration; explainability of security outcomes
- Threat and privacy modeling applied to UX (STRIDE, LINDDUN)
- Operator dashboards for GPU, storage, bandwidth participation; resource scheduling
- Observability of decentralized systems: reliability, rewards, faults, latency
- Incentive design surfaced in UI; failure-mode communication and recovery paths
- AI-assisted accessibility audits and usability clustering for large datasets
- Privacy-preserving ML: federated learning, differential privacy, zkML
- Prototypes for rapid deployment with minimal adoption barriers
- Evaluation of user trust, explainability, and transparency in AI-driven systems
- Studies: protocols, instruments, sanitized or synthetic datasets, analysis notebooks
- Prototypes: minimal working systems that expose interaction patterns before scale
- Tooling: auditors, linters, collectors, visualization components, test harnesses
- Curations: “awesome” lists, standards mappings, reference implementations
- Empirical design: task analyses, cognitive walkthroughs, SUS, NASA-TLX, time-on-task, error rate; inter-rater agreement (Cohen’s kappa)
- Pre-registered protocols where relevant; A/B and counterfactual evaluation
- Reproducibility: deterministic seeds, pinned environments (
requirements.txt,package-lock.json), container specs (Dockerfile), data schemas - Security and privacy alignment: NIST SP 800-63/53, HIPAA 45 CFR Part 164, GDPR Art. 5/25, ISO 9241-210, IEC 62366, WCAG 2.2
- Adoption metric focus: install → first insight (TTFI) in minutes
ux-methods— study templates, consent language, instruments, analysis notebooksai-ux-auditor— AI-assisted accessibility and usability audit CLI with report artifactshealthcare-ux— patient and clinician workflow prototypes with safety checks and logssecurity-ux— key, MFA, and consent flows; error-proofing, recovery, threat modelsdepin-ux— operator dashboards, schedulers, reliability and reward visualizationsai-research-demos— federated learning, differential privacy, zkML examples with UX harnessesawesome-accessible-ai— curated research, standards, and tooling
- Code: MIT by default. Docs and data: CC BY 4.0. Models: Apache-2.0 unless noted
- Each repo declares dataset license, schema, provenance, and limitations
- DOIs via Zenodo for citable releases; semantic versioning for artifacts
- Evidence over assertion; publish limitations and negative results
- No sensitive data; PHI and PII prohibited. Use de-identification or synthetic generation
- Issues and PRs reference tasks, metrics, and standards mappings