Paysight, a SaaS platform to make payroll control reliable

Human Sight publishes Paysight, a web platform that centralises and checks companies' payroll data.

Automatic detection of payroll anomalies
Automated exchanges with the CPAM and Net-Entreprises
A multi-tenant platform evolving since 2023

Centralising payroll to control it better

Human Sight, publisher of the Paysight platform

Since 2023, TCM (The Coding Machine), a custom software development agency based in Paris, has been designing and evolving Paysight, the SaaS payroll control platform.

Human Sight publishes Paysight, a fully web-based, multi-tenant platform. It centralises and checks companies’ payroll data: payslips, pay items, contributions and contribution bases.

The platform also manages daily sickness allowances (IJSS), thanks to automated exchanges with the CPAM (French health insurance) and Net-Entreprises.

In 2023, the starting point was clear: a fully web-based solution able to manage pay items, with an anomaly control module. Since then, the project has progressed through maintenance cycles and successive batches of enhancements.

Three challenges at the heart of payroll control

Business rules with no margin for error

Payroll control is both a legal and an accounting issue. Every business rule had to be perfectly reliable.

Complex, heterogeneous data

The tool had to import, read and analyse payroll data that is often complex and very different from one source to another.

A platform shared by several clients

Paysight is multi-tenant: each company's payroll data had to remain isolated on a shared platform.

A SaaS infrastructure designed to evolve

From rule configuration to anomaly correction

TCM built a complete software infrastructure hosting a modular SaaS application.

A back office is used to configure the legal rules. CSV files go through an asynchronous processing pipeline. Anomalies are then surfaced in interactive dashboards, where they are handled.

Web business application

Reactive front end and back end of the platform.

Payroll control algorithmic engine

Asynchronous workers run the checks on payroll data.

Anomaly analytics dashboards

Detected anomalies are displayed in interactive dashboards so they can be corrected.

3 years
of continuous collaboration, from initial development to evolutive maintenance
4 separate deliverables deployed
6 million data items processed

In payroll, a badly coded rule becomes an error on an employee's payslip. So we built Paysight so that every check is configurable, and we have kept enriching it with each release for three years.

Samar Kallel - Project manager at TCM

SOM
Sonia Mnif
Directrice de projet
Management Conseil Projets complexes IA
Mon objectif quotidien : faire en sorte que chaque ligne ajoutée aujourd'hui facilite notre travail de demain.
10 ans d'expérience
+ 30 projets livrés
SAK
Samar KALLEL
Cheffe de Projet
Projets Web Projet Mobile Conseil
Mon rôle, c’est de faire avancer les projets simplement, sans friction inutile.
9 ans d'expérience
+ 30 projets livrés

Your questions about automated payroll control

By centralising payroll files in a single tool, then applying configurable control rules to each pay item, contribution and contribution base. This is what TCM built for Human Sight with Paysight: CSV files are imported, analysed by an asynchronous engine, and anomalies are surfaced in dashboards to be corrected.

With a multi-tenant architecture that isolates each company’s data while sharing the same application foundation. On Paysight, TCM relies on Laravel and Laravel Tenancy to run a single platform for several clients.

TCM (The Coding Machine), a custom software development agency based in Paris, has been designing and evolving Paysight, the SaaS payroll control platform of Human Sight, since 2023.
The platform also manages daily sickness allowances (IJSS) thanks to automated exchanges with the CPAM and Net-Entreprises.

Business rules that are complex to make reliable?