LNMIIT Recruitment Portal.

01 / overview

Software for a real institutional process.

During the LNMIIT LUSIP internship, I built a faculty recruitment platform around candidate applications, academic records, administrator review, and document output.

The institutional portal link opens an LNMIIT sign-in page; the recruitment workflow is not publicly browsable without access. The implementation source remains private, and this case study covers the work documented in my professional profile and résumé.

02 / problem

Reduce repeated entry without losing reviewability.

Academic applications combine résumé data, publications, experience, permissions, and formal documents. The workflow needs to reduce repetitive data entry while keeping the resulting records available for review.

03 / architecture

Parse, verify, review, and render.

The documented workflow extracts candidate information from résumés, structures academic fields, and integrates Crossref and Scimago data for publication checks.

Role-based access separates candidate and administrator workflows. Document rendering turns application data into consistent PDF output. The diagram summarizes this documented workflow rather than exposing confidential deployment details.

04 / Engineering decisions

Use deterministic checks for academic metadata.

External academic data provides a more inspectable verification path than asking a language model to assert a journal’s indexing or quartile. These checks still depend on the coverage and freshness of the external data.

Extraction and verification serve different purposes: one structures an input, while the other compares specific fields against a reference source.

05 / challenges

Messy documents meet formal records.

Unstructured résumés vary in layout and terminology. Recruitment data also crosses permissions and document-format constraints. The platform joins these concerns into candidate and administrator workflows.

06 / results

Delivered during the LUSIP internship.

I delivered the recruitment platform with résumé parsing, publication verification, role-based permissions, and PDF generation during the June–July 2026 internship. The work connected candidate data entry to institutional review and document output.

07 / learnings

The workflow is part of the architecture.

A useful institutional product needs data extraction, review, permissions, and final documents to work together. The quality of the handoffs matters as much as any individual AI component.