An empathetic product designer, a curious design researcher, a violinist and a writer — from Iran, living in Seattle.
With a B.S. in Computer Science from Shahid Beheshti University and an MHCI+D — a Master of Human-Computer Interaction + Design — from the University of Washington, I bring an engineering perspective and a creative problem-solving approach to design. Beyond AI, visual, and interaction design, my areas of expertise include ethical design, feminist HCI, and voice UX.
A few things I'm proud of: Passbook by Remitly was recognized in Time Magazine's 100 Best Inventions of 2020, and Docufy and Legal Ease were featured at Microsoft Ignite 2023. In 2025 I spoke at the MSD Impact Series on AI UX and responsible-AI design frameworks. Outside work, I mentor with ADPList and served as Community Engagement Director at IxDA Seattle.
The human-facing layer around autonomous systems: how a long run shows progress, how someone steers it mid-flight, and where a person must stay in the loop — placed so agents can run as autonomously as is safe.
Red-teaming and evals for non-deterministic systems — transparency, override paths, and role-based constraint.
AI-simulated personas that pressure-test agent behavior pre-launch, reaching edge cases recruiting can't surface.
I build the agents and AI tools my design work runs on — MCP servers, Copilot Studio and Claude skills, evaluation agents, React/TypeScript prototypes. Built, not specced.
I lead design of the agentic experience layer in a security product at Microsoft — progress on long-running work, reversible actions, mid-flight steering, and role-based access, so security teams can safely give agents more autonomy. Previously (2021–2024) I led end-to-end design strategy for Docufy, ExpenseGo, and Legal Ease. I manage and mentor a team of 4 designers and researchers.
I designed and launched Passbook, an AI-assisted tri-modal banking app recognized in Time Magazine's 100 Best Inventions of 2020, driving 67% activation within the first two months and reporting to the VP of Product.
I optimized the B2B learner progress-monitoring experience by 40%, delivering final designs a month ahead of schedule.
I designed and developed an AI-assisted grading tool for Spanish writing assessments, reaching 92% scoring accuracy vs. 88% average human-evaluator agreement across a 4-school pilot.
Design systems for an educational strategy game, and an automated usability-testing platform evaluating accessibility of Farsi-language sites for elderly users.
More platforms I designed — condensed, since full case studies are under NDA.
Teams + mobile Copilot for expense reporting, MVP shipped to 500 users.
OCR + LLM receipt matching cut manual review ~5 min/report · ~50% auto-validatedAI routing & auto-tagging across litigation hold workflows.
100K+ docs auto-tagged · review time −20% · <3s retrieval · SUS 90/86 · est. $475K+ & 1,000+ hrs saved annuallyRegistration flow rebuilt through heuristic evaluation and usability testing.
Completion 43% → 76% · time 18.2 → 8 minutesLive modelPowered by GPT-4o Grounded in Sayena's case studies, not a canned bot. Billed per question — so make them good ones.