Case Study

Pulse

A self-hosted Professional Intelligence Dashboard that transforms a constant stream of industry updates into a focused, categorized daily briefing.

Project Type Professional Intelligence Dashboard
Role Solution Architect + Builder
Environment Self-Hosted Homelab
Focus Automation + Infrastructure

Overview

A practical system for finding what matters without checking dozens of sources.

Pulse is a self-hosted Professional Intelligence Dashboard designed to answer one question: what changed today that is actually worth my attention?

The platform collects updates from curated industry sources, normalizes the incoming data, removes duplicate items, assigns each story to a professional discipline, and publishes a concise summary to a dashboard.

The Need

Useful information was scattered across too many places.

My work spans commerce, design, web development, print production, automation, and infrastructure. Keeping up with those disciplines meant repeatedly checking blogs, release notes, vendor announcements, publications, and technical feeds.

The challenge was not access to information. It was filtering the volume, reducing repetition, and presenting the most relevant updates in a format that could be reviewed quickly.

My Role

Plan the architecture, deploy the platform, and build the processing workflow.

01

System Architecture

Defined the platform structure, service responsibilities, data flow, hosting environment, storage paths, and dashboard output.

02

Workflow Automation

Built the n8n workflow used to authenticate, retrieve, normalize, filter, categorize, deduplicate, summarize, and export article data.

03

Deployment + Integration

Deployed the containerized services, configured persistent storage, connected the output to the dashboard, and tested the full processing path.

Architecture

Each service handles a specific part of the intelligence pipeline.

  • FreshRSS collects and archives curated source feeds
  • n8n processes articles through a scheduled automation workflow
  • PostgreSQL supports reliable workflow data storage
  • JavaScript transformation steps normalize incoming article data
  • Date filtering limits the summary to current, relevant items
  • Category rules assign stories to professional disciplines
  • Duplicate removal prevents repeated links from crowding the briefing
  • JSON output provides a clean interface for dashboard presentation
  • Homepage displays article counts, category totals, and the leading story

Processing Workflow

From raw feeds to a structured daily summary.

Manual and scheduled triggers begin the workflow. Pulse authenticates with FreshRSS, retrieves the latest articles, splits them into individual records, and standardizes fields such as title, source, publication date, link, and category.

The workflow then filters for current items, assigns disciplines, removes duplicates, builds the summary object, converts the result to JSON, and writes the finished output to persistent storage for the dashboard to read.

Coverage

Built around the disciplines that shape my work.

  • Commerce
  • Design & Branding
  • Web Development
  • AI & Emerging Technology
  • Print & Production
  • Infrastructure & Self-Hosting
  • Business & Professional Development
  • Reading Queue

Technical Stack

A containerized, maintainable platform built from practical open tools.

  • Ubuntu Server
  • Docker Compose
  • FreshRSS
  • n8n
  • PostgreSQL
  • JavaScript
  • JSON
  • Homepage
  • Linux administration and persistent-volume management

Outcome

A working intelligence system instead of another collection of unread links.

Pulse now runs as a live, scheduled platform on my own infrastructure. It consolidates curated sources, processes new content automatically, and presents a readable summary through the same dashboard I use to monitor the rest of my environment.

For my portfolio, Pulse demonstrates a different side of my work: systems thinking, workflow automation, data transformation, container deployment, interface planning, and the ability to turn a loosely defined problem into a dependable working tool.

Next Phase

Designed to grow beyond collection and categorization.

Future development can add ranking, stronger relevance scoring, improved category balancing, historical trend views, source-quality controls, and optional AI-assisted analysis.

The architecture keeps collection, processing, storage, and presentation separate so each layer can evolve without rebuilding the entire platform.

Next

Need a workflow or internal system that turns information into action?

Available for automation, dashboard, commerce, brand-system, and creative-operations projects.