Hannah B. Pasandi

Human Centered and Accessible AI

Accessible and relational AI systems that keep sensitive interaction local, support communication, and reason about human-AI relationships beyond tool use.

Human Centered and Accessible AI

SignSense

Private edge sign language translation.
92.5% sign to gloss accuracy80.4 ms latencylocal processing

SignSense combines a lightweight vision model with an edge LLM to translate sign language locally. It supports real time accessible communication while keeping biometric signing data on device.

translate signing locally, biometrics stay on devicesignervisionlightweightedge LLMtexton-device · 80 ms · 92.5% sign-to-gloss
Human Centered and Accessible AI

Sign Language Atlas

A system map for real time sign language translation.
30 year mapedge privacy lensAR research gaps

Sign Language Atlas organizes three decades of sign language recognition and translation systems across sensing, vision, neural models, privacy, and edge deployment.

three decades of systems, mappedsensingvisionneural modelsprivacyedge deployment1990stoday
Human Centered and Accessible AI

Relate

Relational ethics for human AI interaction.
relational impact assessmentcompanion AI casesgovernance framework

Relate studies moral consideration through relational capacity and embodied interaction rather than unverifiable claims about machine consciousness. It connects posthumanist theory to AI governance.

moral standing from relation, not claimed consciousnesshumanAIrelational capacityembodied interactiongovernancerelational impact assessment