Private opportunity intelligence

Turn scattered demand into a private qualified signal stream.

Opportunity Radar is a self-hosted workflow and dashboard product for finding real opportunities faster. It discovers promising sources, filters noisy posts, validates employer intent, and routes the best matches into a private operator workspace built for action instead of browsing.

Cadence

3h

Live monitoring loop for validated sources

Routing

2+

Telegram and structured review destinations

Shape

Dual

Public-safe story, private operator dashboard

Signal console

Workflow preview

private

Source intake

Reddit RSSFounder communitiesJob boards

Qualification engine

Keyword and intent scoring
Confidence aging and source health
AI job validation
Action-aware routing

Destinations

Telegram alerts

Fast response for top matches

Review workspace

Human triage and history

Workflow overview

One workflow, two radar loops, one private system.

The current implementation is already more than a scraper. It discovers strong communities, monitors them repeatedly, classifies actual opportunity intent, and prepares everything for a more serious data-backed app.

Stage 1

Discover

Discover

Weekly RSS discovery hunts for new high-value communities using curated intent and niche keywords.

Stage 2

Score

Score

Rules scoring measures hiring intent, pain signals, build relevance, freshness, and quality before expensive checks.

Stage 3

Validate

Validate

AI classification only steps in after the first pass strips noise, self-promo, and weak speculative posts.

Stage 4

Route

Route

Urgent matches move to alerts while the rest become structured records inside a private operator workspace.

System structure

Designed as a private operations surface, not a public feed.

The landing page exists to explain the machine. The protected app exists to work with real opportunities. That boundary is part of the product design, not a temporary decision.

Discovery radar

Finds candidate communities from keyword-driven Reddit searches, then keeps only the sources that continue producing signal.

Monitoring radar

Revisits validated sources every three hours, cleans raw posts, and extracts the most actionable opportunities.

Private review layer

Holds the real opportunities, score breakdowns, review actions, run history, and owner-managed access.

Infrastructure

Built around the stack you already run.

Automation core

n8n on Docker

Schedules, scraping, branching, AI classification, and routing live in the workflow engine you already run.

Edge and delivery

Nginx on Oracle VPS

Handles the app entry point cleanly while keeping your self-hosted setup aligned with the rest of the stack.

Storage path

Postgres-first direction

Google Sheets works for transition, but the product is being shaped around structured relational records.

Alerting

Telegram and review queue

High-confidence results can trigger fast action without exposing the full opportunity stream publicly.

Roadmap

Built for the next layer, not just the current feed.

The Reddit workflow is version one of the source engine. The real target is a normalized intelligence system where new adapters can plug in without changing how the dashboard thinks.

More source adapters beyond Reddit
Cross-source deduplication by fingerprint
Source reputation and freshness weighting
Explainable score breakdown inside each lead
Owner-controlled access for trusted collaborators
A dense premium dashboard instead of a generic admin panel

Access model

Open homepage. Closed intelligence layer.

The experience stays visible, but the opportunities stay private. Approved users can log in, review leads, inspect scoring, and monitor runs. Everyone else only sees the system story and request path.