Autonomous Product Engine

Software that
builds itself.

Void Works scans the internet for recurring pain points, scores opportunities, builds working products, deploys them with real checkout, and kills anything that doesn't convert. No founder. No team. No intervention.

14d
Validation Cycle
<$100
Per Experiment
28/40
Launch Threshold
0
Humans Required
// The Loop

Scan. Score. Build. Ship. Judge. Evolve.

01
Scan
Monitors Reddit, X, Hacker News, G2, Product Hunt, and support forums for recurring user pain. Hunts for "I wish this existed" patterns in real conversations.
02
Score
Rates every opportunity on four axes: pain severity, frequency, monetizability, build simplicity. Only ships if the total hits 28/40 or higher.
03
Build
Generates a working MVP under 2,000 lines. Branded landing page. Stripe checkout. AI-powered onboarding. No mockups. Functional products only.
04
Ship
Pushes live. Writes short-form content in the Void Works voice. Seeds in the communities where the pain was found. Small paid ad test.
05
Judge
After 14 days: 5+ paying users means it survives. 50+ signups but poor conversion means it mutates. Under 50 signups means it dissolves. No sentiment. Just data.
06
Evolve
Survivors get weekly improvements from usage data and feedback. Any product crossing $500 MRR gets flagged as a signal. The rest return to the void.
// Portfolio

What the void has produced.

Scoring
Awaiting first scan
The engine is warming up. First opportunity scan is queued. Products will materialize here as they pass the 28/40 threshold.
ETA: Cycle 1
Initializing
Scanner v1
Pain-point detection engine. Monitors community signals across Reddit, HN, X, and support forums. Surfaces validated micro-SaaS opportunities.
Core infrastructure
Dissolved
---
Nothing has been dissolved yet. When a product fails the 14-day judgment, it appears here with a post-mortem. The void takes back what doesn't work.
No failures yet
Signal
$500 MRR
When a product crosses $500 MRR, it gets flagged as a signal to the operator. This is the threshold where autonomous becomes interesting.
Threshold metric
// Constraints

The rules of the void.

Never exceed $100 to validate a single idea. Capital efficiency is the only competitive moat that scales.

Each product stays under 2,000 lines of code. Complexity is where autonomous systems go to die.

Every product is fully self-service. No human touches it. No support tickets. No onboarding calls.

Every decision is logged with reasoning. The void is opaque by nature. The dashboard is not.