Spandan Ghose Chowdhury

—

Unclaimed · AI-structured profileOpenAlex

Early-career researcher bridging quasi-experimental IS methodology and agentic AI for LLM-mediated e-commerce

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-16) 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 scoring is not yet computed. OpenAlex records 5 works in 2026, but venue types and peer-review status are unresolved. This axis would measure cumulative publication volume and yearly productivity across indexed sources.

Impact—

Impact data is insufficient: h-index and i10-index both read 0 with no citation accumulation. This axis would measure citation-based influence — total citations, citing works, and per-paper attention — not yet scorable for this early-stage profile.

Focus0/100

The focus axis scores 0, reflecting a field distribution split exactly 50/50 between research methodology and econometrics in information systems and AI for information retrieval and e-commerce decision support. Topic counts reinforce this breadth: five topics each appear once, spanning neural network applications, recommender systems, Bayesian inference, search behavior, and multi-agent negotiation. The profile currently reads as deliberately dual-track rather than specialized within a single field.

Collaboration0/100

The collaboration axis scores 0 because the underlying record captures 0 distinct co-authors, and the frequent-coauthor list is empty. Both reviewed 2026 arXiv papers list the author in the first and corresponding position with no other captured names, suggesting current catalog work is single-authored. As more indexed output accumulates, this axis would reveal co-authorship networks and team science patterns that the present snapshot cannot yet show.

Activity52/100

The activity axis scores 52, a moderate level consistent with a concentrated recent burst: OpenAlex records 5 works in the last three years, all of them falling in 2026. Nothing is indexed before that year, so the score reflects short, current momentum rather than a sustained multi-year cadence. Both reviewed papers appeared on arXiv in 2026, reinforcing that picture of an active, newly visible research phase.

Reproducibility60/100

The reproducibility axis scores 60, computed as the mean readiness across the 2 papers Evolve has assessed. The spread is wide: the Monte Carlo power-calibration study is rated high readiness, fitting its simulation-based design, while the agentic share-of-search system is rated low because its key evidence is synthetic with only a tiny live pilot. The average masks this contrast between computational and system-building work.

Research summary

Spandan Ghose Chowdhury is an early-stage researcher whose recorded work sits at the intersection of two fields, splitting evenly 50/50 between research methodology and econometrics in information systems and artificial intelligence for information retrieval and e-commerce decision support. Evolve has reviewed two 2026 arXiv papers, both listing the author in the first and corresponding position. The first, 'Beyond the Illusion of Power,' delivers a 9,837-cell Monte Carlo decomposition of power gaps between difference-in-differences and instrumental-variable designs in observational IS panels, including staggered adoption; Evolve judges it rigorous and unusually self-critical, though simplified estimators temper its claims, and rates its reproducibility readiness high. The second, 'Agentic Share-of-Search,' proposes an ASoS metric and a ReAct-based multi-agent diagnostic system for competitive decision-making in LLM-mediated retail; its evidence remains largely synthetic with a tiny live pilot, so reproducibility readiness is low. OpenAlex records 5 works, all dated 2026, with h-index 0 and i10-index 0 and 0 distinct co-authors captured, so impact and collaboration signals remain sparse. The reproducibility axis averages 60 across the 2 assessed papers and the activity axis scores 52. This profile is name-resolved from public bibliographic data and is best read as an early snapshot of a methods-driven, fast-moving agenda.

Collaboration pattern

The current record shows a strongly solo pattern: OpenAlex captures 0 distinct co-authors, the frequent-coauthor list is empty, and both reviewed 2026 arXiv papers place the author in the first and corresponding position alone. The collaboration dimension consequently scores 0. This fits an early-stage, methods-driven agenda where simulation studies and agent prototypes can be executed independently, though future indexed work may reveal collaborators not yet visible in this snapshot.

Research keywords

power analysisdifference-in-differencesinstrumental variablesMonte Carlo simulationpanel datastaggered adoptionAgentic Share-of-Searchmulti-agent systemsLLM shopping assistantsgenerative engine optimizationroot cause analysisReAct

Highlights

A 9,837-cell Monte Carlo audit of quasi-experimental power in IS panels
The first-author, corresponding paper 'Beyond the Illusion of Power' decomposes power gaps between difference-in-differences and instrumental-variable designs across 9,837 Monte Carlo cells in observational IS panels, covering staggered adoption settings. Evolve's verdict calls it rigorous and unusually self-critical, while noting that simplified estimators temper some claims; reproducibility readiness is rated high.
Agentic Share-of-Search: measuring visibility in LLM-mediated retail
The second first-author, corresponding paper proposes the ASoS metric plus a ReAct-based diagnostic multi-agent system for competitive decision-making where LLM shopping assistants mediate e-commerce. Evolve credits its honest feasibility framing but flags that key evidence is synthetic with only a tiny live pilot, yielding low reproducibility readiness.
Reproducibility contrast: simulation-first rigor vs. system prototypes
Across the 2 assessed papers the reproducibility mean is 60, but the two ratings diverge sharply — high for the simulation-based power study, low for the agentic system. The pattern suggests the author's methods-first work reruns more easily than early-stage AI system prototypes, a useful signal for anyone planning to build on either line of research.

Timeline

2026Posts 'Beyond the Illusion of Power: Calibrating Quasi-Experiments in Observational IS' to arXiv as first and corresponding author.
2026Posts 'Agentic Share-of-Search: A Multi-Agent AI System for Competitive Decision-Making in LLM-Mediated E-Commerce' to arXiv as first and corresponding author.
2026OpenAlex records the first indexed works on file: 5 publications, all dated 2026, the only year with any recorded activity.

Papers in the Evolve catalog

2026-09-23first authorcorresponding
Beyond the Illusion of Power: Calibrating Quasi-Experiments in Observational IS
2026-09-10first authorcorresponding
Agentic Share-of-Search: A Multi-Agent AI System for Competitive Decision-Making in LLM-Mediated E-Commerce

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.

All bibliometrics come from OpenAlex only (identity: openalex); no institution, country, or ORCID is on record.
Citation-impact measures are empty (h-index 0, i10-index 0), and the output and impact radar axes lack sufficient data to score.
Detailed coverage extends only to the 2 papers Evolve has reviewed; OpenAlex lists 5 works in total for 2026.
This profile is name-resolved from public bibliographic data; the name may be shared by multiple individuals.