[A :: M]
Autonomous Experiment Lab
Surfaces how your systems really behave — workflows, failures, and hidden opportunities — in real time.
Quick Diagnostic
Tell us what's breaking
Three questions. One AI-powered preliminary diagnosis. No commit needed.
Operator Feed
Live console output
Every event — deployments, telemetry, experiment changes — streams through the Operator in real time.
Live Proof
Verifiable system activity
No marketing claims. These metrics pull directly from the lab's D1 telemetry. Cached at edge, refreshed every minute.
The Model
Five engines, one operation
Cash work funds experiments. Experiments become products. Everything documented publicly. Nothing invented.
Active Experiments
Three tests running right now
These are the active experiments. Each has a published hypothesis, current stage, and kill condition. Revenue numbers appear when revenue exists.
Participate
Ways to engage
Each path maps to a specific experiment or engine. Pick what fits.
Audit or Sprint
Book an Automation Audit ($25/$300) or a Revenue Sprint ($500). We fix broken workflows or build new ones.
Products
Purchase automation templates, prompt packs, and datasets — only after experiments validate demand.
Field Notes
Subscribe to the newsletter. Get each experiment's outcome — including failures — delivered weekly.
Propose an Experiment
Have an idea for an AI income experiment we should run? Suggest it. If it passes validation, we'll build it.
Join the Team
Operator feed, experiment updates, behind-the-scenes logs, and community Discord. $1/mo on Gumroad.
Experiment Backlog
Queued hypotheses
Ideas waiting for bandwidth or a trigger condition. None are active. None have invented revenue. They move into active rotation when an experiment slot opens.
| ID | Hypothesis | Engine | Trigger |
|---|---|---|---|
| B001 | Cold Diagnosis Outreach | Experiment | Bandwidth opens |
| B002 | AI Chatbot Setup | Experiment | 3 presales first warming |
| B003 | Data Extraction Service | Experiment | 2 presales first warming |
| B004 | Big Board Sponsorships | Opportunity | 100+ visitors/mo |
| B006 | Diagnosis Product Pipeline | Product | 10 subs/month |
| B007 | Automation Template Store | Product | 500 subs |
| B008 | Tool Affiliate Funnel | Opportunity | Content lib |
| B009 | Freelancer Lead Matching | Opportunity | Traffic cap |
| B010 | Niche Prompt Library | Experiment | 1 niche warming |
| B011 | LinkedIn Post Generator | Experiment | 3 presales warming |
FAQ
Common questions
A public experimental lab testing ways to make money with AI. We run small experiments, publish the results (including failures), and sell only what generates verified demand.
Transparency builds trust. We don't invent revenue or pretend experiments succeeded. Publishing failures is how the audience learns — and how we stay honest.
Not yet. The cash engine (audits and repair) is launched and taking signups. Revenue will be posted publicly the moment it arrives.
A consultancy sells expertise. We sell evidence. Experiments either produce revenue or get killed — and you can see which happened, and why.
Zapier, Make, Airtable, Notion, Typeform, Gravity Forms, Webflow, HubSpot, Salesforce, n8n, ActiveCampaign, and Slack.
Each experiment has a published kill condition (e.g., "zero presales from 50 outreaches"). When that condition is met, the experiment stops. Results published regardless.
No invented revenue. Just results.
We run experiments publicly. When they produce revenue, you'll see the number. When they fail, you'll see that too.
Book an Audit