Project snapshot - Humanitarian Jobs Dashboard

Project Overview

The Goal

To build a self-hosted, sovereign data platform that tracks and analyzes the humanitarian labor market by monitoring the ReliefWeb API. The project aims to identify shifts in sectoral demand (e.g., Wash vs. Protection) and geographic hiring trends to help professionals navigate the sector.

The Solution

A comprehensive “Humanitarian Labor Intelligence” dashboard. It uses n8n to poll the ReliefWeb /jobs endpoint, Flowise to perform semantic analysis on job descriptions (detecting emerging “soft skill” requirements), and a Nuxt frontend to visualize these insights—all containerized and self-hosted for privacy and cost-efficiency.

Key Objectives

• Automated Data Ingestion: Use the ReliefWeb API to fetch new vacancies daily, including metadata like Career Category, Years of Experience, and Theme.

• Skill Gap Analysis: Leverage Flowise (AI) to extract specific technical requirements (e.g., “SQL,” “KoboToolbox,” “PowerBI”) from unstructured job descriptions that standard filters miss.

• Privacy-First Self-Hosting: Deploy the entire stack using Docker, ensuring no third-party client data is shared and maintaining a permanent, private historical archive of job trends.

Audience & Stakeholders

• Primary Users: Humanitarian workers (job seekers), IM practitioners, and HR researchers.

• Key Stakeholders: None (Personal Portfolio Project).

The Plan & Key Features

Overall Approach

The project utilizes a Modular Self-Hosted Architecture.

1. n8n triggers every 12 hours, querying https://api.reliefweb.int/v2/jobs.

2. Data is cleaned and sent to a local database.

3. Flowise processes the text to categorize the “seniority” and “technical intensity” of the roles.

4. Nuxt serves as the presentation layer, pulling from the database to show real-time charts.

Core Components

• API Integrator (n8n): Handles the logic for the ReliefWeb API, including pagination and filtering for specific “Themes” (e.g., Coordination or Food Security).

• Trend Engine (Flowise): A self-hosted RAG (Retrieval-Augmented Generation) pipeline that lets you “chat” with the current job market (e.g., “What is the most common skill requested for GIS roles in 2026?”).

• The Dashboard (Nuxt): A fast, responsive UI designed in Figma, featuring high-density data visualizations like “Hiring Heatmaps” and “Skill Word Clouds.”

• Deployment (Docker): A single docker-compose.yml file managing the Nuxt app, n8n instance, and Flowise environment.

Timeline & Deliverables

Major Milestones

• Phase 1: API & Design: Map out the ReliefWeb JSON structure and design the Dashboard UI in Figma.

• Phase 2: Data Pipeline: Build the n8n workflow to fetch and store job data (using the appname parameter required by ReliefWeb).

• Phase 3: AI Categorization: Set up Flowise to analyze job descriptions for hidden technical requirements.

• Phase 4: Dashboard Build: Develop the Nuxt frontend and integrate it with the self-hosted backend.

Final Deliverables

• Self-Hosted URL: A live link to the personal dashboard instance.

• n8n Workflow (JSON): The automated pipeline for ReliefWeb data extraction.

• Technical Documentation: A “beginner-friendly” guide on how to deploy this exact humanitarian IM stack using Docker.

• Figma File: Access to the UI/UX design components used for the dashboard.

---

A Technical Note for your n8n workflow:

ReliefWeb’s API is very friendly but requires an appname parameter in every request. Since you are a beginner, the URL you’ll likely use in your n8n HTTP Request node will look like this:

https://api.reliefweb.int/v2/jobs?appname=my-humanitarian-dashboard&limit=100&preset=latest