01Work02Services03About04Writing05Contact

amansonwani43@gmail.com

← All work

Placement Management System

A campus-wide platform that turned chaotic placement season into a single, calm workflow — and became a published research paper.

Role

Full-stack developer & researcher — I designed the UX in Figma, modelled the database, built both frontend and backend, and co-authored the research paper.

Timeline

4 months · Solo build with faculty guidance

Stack
  • React
  • Node.js
  • Express
  • PostgreSQL
  • TailwindCSS
  • JWT Auth
Year

2024

The problem

Placement season at our institute ran on spreadsheets, notice boards and WhatsApp groups. Students missed deadlines because announcements were scattered, admins re-entered the same data in multiple sheets, and companies had no single view of applicants.

The placement cell estimated hours per week went into manually matching student eligibility (CGPA, backlogs, branch) against each company's criteria — work a database query should do in milliseconds.

My approach

(01)

Research before pixels

Interviewed placement officers, students and a recruiting company to map the actual workflow — then documented it as the baseline for the research paper.

(02)

Model the data first

Designed the ERD around three actors (students, admins, companies) with eligibility rules as queryable data, not documents — so matching became a WHERE clause instead of an afternoon of Excel.

(03)

Role-based dashboards

Designed three focused dashboards in Figma: students see only relevant drives and their status; admins manage drives and vacancies; companies see pre-filtered eligible applicants.

(04)

Build, test, publish

Built the React + Node.js implementation, ran it against real placement-cell scenarios, and wrote up the system design and outcomes for publication.

The solution

A single web platform where admins create placement drives with structured eligibility criteria, students apply with one click from a personalised feed, and every CV, status change and offer lives in one place.

Automated eligibility filtering removed the manual matching step entirely — when a company sets 'CGPA ≥ 7, no active backlogs, CS/IT only', the applicant list is already correct.

The outcome

1

Research paper published

3

Role-based dashboards

~80%

Less manual admin matching

Resource Management System