Skip to content
Antino
All case studies

Marketplaces · Expert Marketplace Platform

1lattice

Multi-sided B2B marketplace platform (Python-Django backend + React frontend) connecting expert/professional service.

1lattice
25
Weeks operated
2k+
Hours of work
3
Engineers

The brief

Build a production-grade multi-sided marketplace platform enabling corporate clients (customers) to post projects, receive structured responses from verified expert/professional users, manage contracts and payments, and be administered through a configurable internal panel — with data intelligence fed from scraped e-commerce and healthcare datasets.

What we built

A full-stack, multi-panel expert marketplace and project management platform built on Python-Django REST API and React. The system includes:

  • four distinct user-facing panels — customer portal, expert/professional panel, internal admin panel, and iPanel
  • complete project lifecycle workflow engine from creation through expert response, client approval, and auto-generated invoice
  • full finance stack with PDF invoices, payment methods (bank/UPI/PayPal/PAN), purchase orders, contracts, and addendums
  • configurable survey/questionnaire engine for expert qualification
  • multi-provider authentication (email/OTP, LinkedIn, Google, Facebook)
  • granular RBAC with user types, subtypes, company mappings, and platform middleware
  • a web scraping pipeline covering Meesho, Snapdeal, Amazon, and Flipkart plus a MedIQ healthcare data pipeline; and
  • XLSX report generation and API-level logging. Deployed on AWS Amplify (frontend) with Django backend.
Live in production

The platform is deployed (frontend on AWS Amplify) with all core modules live: multi-panel UI for customers, experts, and admins; project workflow with publish/approve/invoice triggers; full finance suite; social login; team onboarding; survey/questionnaire system; document management; and admin filters. KT and project handover were completed on Feb 21, 2025, indicating a production-ready state at engagement close.

Delivery timeline

How it was built, phase by phase.

8 workstreams across 25 weeks of operated delivery.

  1. buildWeek 1–4 (Sep 9 – Oct 4, 2024)

    Core Platform API Infrastructure

    Foundation-layer Django REST API development covering invoices, payments, contracts, subscriptions, FAQs, and project management — built with CRUD, filters, pagination, and serializer validations from day one.

    Full CRUD API surface covering the core financial and project objects with role-aware listing and filtering

    PythonDjangoDjango REST FrameworkPostgreSQL
  2. buildWeek 2–20 (Sep 17, 2024 – Feb 21, 2025)

    Multi-Panel Frontend (Customer / Expert / Admin / iPanel)

    React-based frontend development covering four distinct user-facing panels — customer portal, expert/professional panel, internal admin panel, and iPanel — each with tailored navigation, forms, and data tables.

    Fully integrated multi-panel application deployed on AWS Amplify with role-aware navigation and data rendering

    ReactContext APIZodAWS AmplifyTailwind CSS
  3. buildWeek 2–20 (Sep 24, 2024 – Jan 17, 2025)

    Identity, Authentication & Social Login

    Comprehensive auth layer spanning OTP-based login, forgot password, email/phone change, LinkedIn/Google/Facebook social login via django-allauth, CSRF handling, and join-link token-based onboarding flows.

    Fully operational multi-provider auth supporting email/OTP, LinkedIn, Google, Facebook, and token-based join-link onboarding

    django-allauthLinkedIn OAuthGoogle OAuthFacebook OAuthOTPJWT/Session
  4. buildWeek 3–22 (Sep 20, 2024 – Feb 14, 2025)

    Role-Based Access Control & Multi-Tenancy

    Granular RBAC system with user types (customer, professional/expert, internal), subtypes, role memberships, company mappings, platform middleware, and permission-check APIs enabling multi-company.

    Platform middleware enforcing role-context across all APIs, supporting customers with multiple company roles and internal admin override capabilities

    PythonDjangoDjango MiddlewarePostgreSQL
  5. integrateWeek 5–9 (Oct 9 – Nov 6, 2024)

    Data Scraping & MedIQ Data Pipeline

    Web scraping of e-commerce platforms (Meesho, Snapdeal, Amazon, Flipkart) for product data, plus a separate MedIQ healthcare data pipeline with custom Django models, upload scripts, and hospital deduplication logic.

    Structured product catalog data from 4 e-commerce platforms and MedIQ hospital/healthcare data seeded into the platform database

    PythonWeb scraping (BeautifulSoup/Scrapy)CSVDjango ORMPostgreSQL
  6. buildWeek 8–14 (Oct 29 – Dec 14, 2024)

    Survey & Questionnaire Engine

    Configurable survey system with question creation, user answer collection, project-linked questions, response tracking with unanswered-question detection, and dynamic question forms in the project admin UI.

    Project-linked survey engine enabling structured expert qualification and response collection with completeness tracking

    PythonDjangoDjango REST FrameworkReact
  7. buildWeek 10–22 (Nov 19, 2024 – Jan 30, 2025)

    Project Workflow Engine

    State-machine-style project lifecycle management with workflow cards, event tracking, tag matching, response collection, client approval, invoice auto-generation on workflow events, and history viewing.

    End-to-end project lifecycle: create → publish → expert response → client approval → invoice generation, all with audit history

    PythonDjangoDjango REST FrameworkPostgreSQLEmail/Notifications
  8. buildWeek 12–22 (Nov 28, 2024 – Feb 7, 2025)

    Finance Module (Invoicing, Payments, Contracts, Addendums, POs)

    Full finance stack including manual and auto-generated PDF invoices, payment method management with OTP verification, PAN card validation, purchase orders, addendums, signed document uploads, and currency conversion.

    Complete finance operations suite: auto-invoice PDF on workflow events, multi-method payments with bank/UPI/PayPal.

    PythonDjangoPDF generationCurrency APIOTPAWS S3 (document storage)

09 · Run a function

Stop renting hours. Start running functions.

Pick the function you want off your plate. We'll map the brain and name the outcome we'd commit to — before you do.