KaamSetu: Design and Implementation of a Location-Based Multilingual Workforce Marketplace for the Informal Sector
Abstract India's informal workforce faces challenges in finding consistent employment, while employers struggle to identify reliable workers for short-term and urgent tasks. This paper presents KaamSetu, a location-based multilingual workforce marketplace designed to connect daily-wage workers with nearby employers. The platform uses a transparent seven-factor weighted matching algorithm based on skill match (35%), location match (20%), experience (15%), salary compatibility (10%), work type (10%), availability (5%), and job category (5%). It supports five languages: English, Hindi, Marathi, Kannada, and Telugu. KaamSetu is implemented using React, TypeScript, FastAPI, and PostgreSQL, with a custom taxonomy of 35 skills across 19 categories. Trust mechanisms include SMS OTP phone verification, administrator-reviewed government ID verification, and user ratings. Implementation verification achieved 153/153 passing automated tests. The matching computation benchmark was approximately 0.0106 ms, while locally measured API latency ranged from 2.08 to 15.85 ms. Matching quality against human-labeled ground truth and user-study evaluation remain future work. Keywords: Informal workforce, daily wage workers, job matching, weighted scoring, geolocation, multilingual interface, trust and verification.
Authors
- Abhijeet Tambe
- Neha Kokate -
- Devang Walke
- Prathamesh Garad
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-18
- DOI
- https://doi.org/10.5281/zenodo.24342867
- Primary Topic
- Digital Economy and Work Transformation
- Type
- preprint