Case Study · Technical Architecture

NeuroSynk Research

We designed and built NeuroSynk Research a single platform to run their entire organization — HR, payroll, finance, projects, internal comms, and a full exam-and-certification engine — replacing a scattered stack of spreadsheets and disconnected tools with one role-based system of record.

See the visual story

System

Enterprise Operations Platform

Research & Professional Services

Scope

15+ modules

55+ data models · 200+ API endpoints · 2 languages

Stack

23 technologies

Across 3 layers

The constraint

What the system had to solve before a line of it was designed.

Like every organization that outgrows its stack, NeuroSynk had hit the point where the software was slowing the work down instead of speeding it up. Their data was fragmented — employee records, attendance, leave, and payroll all lived in separate files, and reconciling them at month-end was manual, slow, and error-prone. There were no real access boundaries — people either had too much access or had to ask someone else to pull a report. Workflows were disconnected: a single leave request, its approval, and its downstream effect on attendance and payroll were three separate manual steps in three separate places. And leadership had no live picture of the business — answering "how are we doing this month?" meant assembling spreadsheets by hand. On top of all of that, NeuroSynk had a need no off-the-shelf HR product could meet: they had to run examinations and issue certificates that outside parties could independently verify.

Architecture

The stack read top down, from what the user touches to what keeps it running in production.

Client

What the user touches — the screens, the app shell, and the state that keeps them responsive.

  • Next.js
  • React
  • TypeScript
  • Tailwind CSS
  • React Hook Form

Services & API

Business rules, authentication, background work, and every third-party system the product talks to.

  • Supabase Auth
  • JWT + bcrypt
  • Radix UI
  • Framer Motion
  • GSAP
  • Lottie
  • TanStack Query
  • TanStack Table
  • FullCalendar
  • dnd-kit
  • ApexCharts
  • Chart.js
  • Recharts
  • Leaflet
  • next-intl
  • Razorpay

Data

The system of record, the indexes that make it searchable, and the stores that hold everything else.

  • MongoDB
  • Mongoose

Modules delivered

Each one shipped production-grade, not as a demo surface.

  • 01People & HR — attendance with punch-in/out, shift scheduling, and live dashboards
  • 02People & HR — leave-and-approval workflows tied directly to attendance
  • 03People & HR — payroll and payslip generation with configurable components
  • 04People & HR — templated HR documents, employee letters, recruitment and internship pipelines
  • 05Finance — transaction views, banking, expense and reporting dashboards, Razorpay payments
  • 06Projects & collaboration — Kanban boards, task management, team calendar, chat, email, meetings, notices
  • 07Examination & certification — question banks, exam definitions, distraction-controlled exam environment
  • 08Examination & certification — automated scoring and publicly verifiable certificate issuance
  • 09Dashboards — role-aware analytics, CRM, project, banking and attendance views
  • 10Platform foundation — granular RBAC, EN/AR multilingual (RTL), notifications, email automations, admin suite

Scope delivered

What the build covered, end to end.

  1. 0155+ data models
  2. 02200+ API endpoints
  3. 032 languages

Design decisions

What we learned before building, and what the built system moved.

We started with discovery, sitting with each department — HR, finance, project leads, and the exam administrators — to map how work actually flowed rather than how the org chart said it should. That map became the module list. From there we designed a single data and permission model. Instead of bolting modules together after the fact, we built one database and one role-based access layer from day one, so attendance, leave, payroll, and finance all reference the same source of truth. We delivered module by module, beginning with the HR core, so the team could adopt the platform in stages rather than through a risky big-bang cutover. Every screen was designed and reviewed against a single component system before it was wired to the backend, which kept the interface coherent across more than fifteen modules. And internationalization was architected in from the start — every label flows through a translation layer, so the platform is multilingual and right-to-left ready for Arabic by design rather than as an afterthought.

NeuroSynk Research now runs its operations on a single platform. The build unified HR, finance, project, and certification workflows that once spanned multiple disconnected tools behind one login and one permission model. It replaced manual, multi-file month-end reconciliation with a single source of truth shared across attendance, leave, and payroll. It connected the leave-request → approval → attendance chain into one automated workflow instead of three manual steps in three places. It gave leadership live operational visibility for the first time — through role-aware dashboards instead of hand-assembled spreadsheets. It unlocked a capability the organization didn't previously have: issuing verifiable credentials that any external party can independently confirm through a public verification page. It enforced least-privilege access across every module, so each role sees exactly what it needs and nothing more. And it made the platform multilingual by design — English and Arabic, fully right-to-left — ready to serve a wider workforce without re-engineering.