Hybrid query-resolution platform routing queries across AI assistant, peer layer, curated FAQ corpus and administrator escalation

Design · All designs
Evidence · 3 claims (3 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 3 claims rest on one study

Description

A deployed support platform for a large online internship programme that routes each raised query across four paths: an AI chatbot answering only from a curated FAQ corpus, a peer layer where any participant may answer, administrator handling, and a scripted facility. The article states that "The platform routes a raised query first to the AI chatbot, which answers only from a curated corpus of verified FAQ entries, a deliberate restriction that prevents unverified procedural guidance". Peer answers pass through administrator acceptance and senior-administrator final approval before release, and the corpus was revised daily so rule or schedule changes were reflected the same day.

Design Implications

Context

Requirements

• A curated corpus of verified FAQ entries that is revised daily so procedural changes are reflected the same day
• Administrator acceptance and senior-administrator approval of peer answers before release to the asker

Constraints

• The AI assistant answers only from the curated corpus and cannot report an individual participant's current status
• The study covers a single programme across nine weeks of onboarding, with volume peaking in week 3 and tapering after, so it describes an onboarding period rather than a steady state

Target Learners

• Participants in a large-scale online internship programme during staggered-cohort onboarding

Learning Goals

• Resolving operational onboarding queries (eligibility, documentation, scheduling, status of pending submissions) with short turnaround without a dedicated mentor per student

Claims

• Status Enquiries Pending Submissions Assistant 1 3 Percent [+M]
• Peer Answering Fast 70 Percent Within One Hour [+M]
• Corpus Coverage 8 Percent Person Resolved Queries [+M]

Related Designs

• Custom AI assistants with curated legal knowledge bases, queried in de-identified form
• Detect-verify-escalate pipeline with four pedagogically distinct routes

Examples

-

Key Sources

• Gupta, J., Ayinampudi, P., Aditya B.M.V., Hegade, P., Sharma, R., Sharma, S., Meenakshi, V., Iyengar, S.R.S. (2026). What Students Actually Ask: Demand Structure and Automation Potential in a Hybrid Support System. arXiv:2609.38967. https://arxiv.org/abs/2609.38967. doi:10.48550/arxiv.2609.38967

Page metadata

Field	Value
type	design
id	hybrid-query-resolution-platform-internship-onboarding
title	Hybrid query-resolution platform routing queries across AI assistant, peer layer, curated FAQ corpus and administrator escalation
description	A deployed support platform for a large online internship programme that routes each raised query across four paths: an AI chatbot answering only from a curated FAQ corpus, a peer layer where any participant may answe...
status	review
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content
sources	1 — gupta-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.