Smit Desai

—

Unclaimed · AI-structured profilename-matched identity

HCI researcher working on AI-assisted decision-making and relational voice agents for older adults

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-12) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Output — total publications, venues, and career productivity — cannot be scored because no full publication list or venue history is available in the underlying data; only the two cataloged arXiv papers are on record.

Impact—

Impact — citations, uptake, and influence on subsequent work — is not yet measurable: the underlying data contains no citation counts or downstream references, so this axis is left unscored rather than estimated.

Focus0/100

The focus score is 0, driven by a perfectly even field distribution: the two cataloged papers split 0.5 to human-computer interaction and 0.5 to HCI (AI-assisted decision-making). With only two papers on record, this reads as a genuinely two-track agenda — relational voice agents for older adults alongside personalization in AI-assisted decisions — rather than a settled specialty; more papers would clarify where the center of gravity lies.

Collaboration44/100

Collaboration scores 44 on a 0–100 scale, built from 8 distinct coauthors across the cataloged papers. The recurring partner is Hasibur Rahman, listed twice in the co-author pool and thus involved in both 2026 papers; the wider circle includes Benjamin R. Cowan, Mahsa Nasri, Manasi Vaidya, Melika Vafafar, and Jessie Chin. Both papers place Smit Desai in the last author position, consistent with a senior or supervisory role, while the moderate score reflects a still-small network.

Activity—

Activity — recent publication cadence and ongoing output rate — cannot be scored: the underlying data lacks a dated publication stream beyond the two 2026 arXiv entries, so there is not enough signal to say whether this pace is sustained.

Reproducibility38/100

Reproducibility readiness averages 38 across the 2 papers scored. The spread maps directly onto Evolve's assessments: the preregistered N=240 personalization experiment rates medium, while the MeBo companion study rates low, with simulated-year vignettes and no baseline cited as limits. A score in the lower-middle of the scale suggests methods are documented but not yet at open-replication standard; preregistration in one study is the strongest signal so far.

Research summary

Smit Desai is a human-computer interaction researcher whose Evolve catalog centers on conversational AI and how people make decisions with intelligent systems. The cataloged record — two arXiv papers, both dated 2026, with Smit Desai in the last author position — splits evenly (0.5/0.5) between human-computer interaction and its AI-assisted decision-making subfield. The first paper, 'MeBo Leaves a Piece of You Behind,' reports a relational voice-based memory companion for older adults, developed through participatory design with 11 older adults and evaluated with 20 seniors, reaching a System Usability Scale score of 87.75, though Evolve notes simulated-year vignettes and the absence of a baseline limit its claims. The second, 'Annie, Are You Okay?', is a preregistered experiment with 240 participants examining how style- and context-based personalization shape AI-assisted decision-making, finding that context-based personalization sways stock choices without changing trust. Bibliometric signals are modest but real: 8 distinct coauthors, a collaboration score of 44, and a mean reproducibility readiness of 38 across 2 scored papers. No institution, country, or ORCID is available, so this profile is name-resolved from public bibliographic data and should be read as a partial, early-stage record rather than a complete career portrait.

Collaboration pattern

Smit Desai's cataloged work is co-authored throughout: 8 distinct coauthors appear across the two 2026 papers, with Hasibur Rahman the only recurring partner (listed twice, so present on both). The broader circle includes Benjamin R. Cowan, Mahsa Nasri, Manasi Vaidya, Melika Vafafar, and Jessie Chin. Consistent last-author placement suggests a coordinating or supervisory role, and the collaboration score of 44 reflects a functioning but still compact network centered on conversational-agent research.

Research keywords

conversational agentsvoice user interfacesolder adultsautobiographical memoryreminiscencerelational agentspersonalized AIpersuasionAI-assisted decision-makingstyle-based personalizationcontext-based personalization

Highlights

Voice memory companion for older adults earns high usability marks
The MeBo project used participatory design with 11 older adults to build a relational voice-based memory companion, then reached a System Usability Scale score of 87.75 in evaluation with 20 seniors — though simulated-year vignettes and the absence of a baseline temper the claims.
Preregistered test of how personalization sways AI-assisted choices
In a preregistered experiment with 240 participants, style- and context-based personalization were tested in AI-assisted decision-making; context-based personalization shifted stock choices while trust stayed unchanged, with minor hypothetical-stakes caveats noted by Evolve.
Consistent senior-author role across a dual-track 2026 agenda
Both cataloged papers place Smit Desai in the last author position across two distinct lines of work — relational voice agents for older adults and personalization in AI-assisted decision-making — with the field distribution split exactly 0.5/0.5 between HCI and its AI-decision subfield.

Timeline

2026'MeBo Leaves a Piece of You Behind' appears on arXiv: a relational voice-based memory companion co-designed with 11 older adults and rated 87.75 on the System Usability Scale by 20 seniors.
2026'Annie, Are You Okay?' appears on arXiv: a preregistered 240-participant experiment showing context-based personalization sways stock choices in AI-assisted decision-making without changing trust.
2026Both cataloged papers list Smit Desai in the last author position, and the year's output splits evenly between human-computer interaction and its AI-assisted decision-making subfield.

Papers in the Evolve catalog

Frequent co-authors

Hasibur Rahman×2 Smit Desai×2 Hasibur Rahman×2 Benjamin R. Cowan×1 Mahsa Nasri×1 Manasi Vaidya×1 Melika Vafafar×1 Jessie Chin×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Bibliometric scores (collaboration 44, focus 0, reproducibility 38) come from Evolve's snapshot of public bibliographic records; no institution, country, or ORCID is available, so identity is resolved by name only.
Output, impact, and activity are unscored because the underlying data has no full publication list, citation counts, or dated publication stream beyond the two cataloged arXiv papers.
Only two arXiv papers (both 2026, both last-author) are in the reviewed catalog; the profile may not reflect the full publication record.
The frequent-coauthor list includes the researcher's own name, likely a deduplication artifact; co-author attributions should be read with that in mind.