Philosophy

AI You Can See

AI is becoming the nervous system of how we build. A nervous system you can't see is one you can't trust. We believe AI should be more transparent than doing it yourself—not less.

The Problem

Most AI assistants disappear into a black box.

"Sure, I'll deploy that!" ...silence... "Done!"

What actually happened? You won't know until something breaks.

We think that's backwards.

Traditional AI
"I'll deploy that for you"
...silence...
"Done!"
What actually happened?
Caveman AI
Opens /services/new
Fills: Name = "gyr", Host = Fly.io
"Which region?" [iad] [sjc] [ams]
Continues with your choice

The Belief

When AI does something, you should see exactly what it's doing. Not because we don't trust AI—because trust requires visibility.

AI should teach, not just do. Watch it once. Do it yourself next time. The goal isn't to replace you. It's to amplify you.

"Humans decide what to build. Everything else adapts."

AI is part of "everything else." It should adapt to you—not replace you invisibly.

No Magic

Everything AI does is visible. No hidden API calls, no secret actions. You see the same forms AI fills.

Same Code Path

AI uses the exact same UI you do. Same validation, same permissions, same everything. No backdoors.

User Learns

Watch AI do it once, do it yourself next time. AI teaches you the system by using it in front of you.

How It Works

Here's what "transparency over magic" looks like in practice.

GHOST MODE: DEPLOYING GYR TO FLY.IO
Step 1: Create Service
Step 2: Environment Variables
DATABASE_URL From Secrets
API_KEY ? Enter value
API_KEY not in your secrets. Enter value or add to secrets?
Step 3: Review & Deploy

You control how much autonomy AI has:

Reactive You ask

AI shows plan, fills forms, you confirm key steps

deploy gyr to fly
Advisory AI notices

AI suggests action, you decide what to do

Memory 40% higher than normal
Autonomous Rule matches

AI acts automatically, you get notified

CPU > 90% for 10 min

The Nervous System

Traditional AI monitoring searches through databases. Caveman AI asks your servers directly—live, real-time, bidirectional.

This is the nervous system we talked about. Not pipelines. Live connection.

Live connection, not pipeline queries

"What's using memory on api-prod-2?" → AI asks the probe → probe checks /proc → real-time answer. No Elasticsearch. No PromQL. Just ask.

Anomaly

Memory 40% higher than normal

Possible leak in api-prod since deploy #847

Root Cause

Latency spike correlates with deploy

Response time increased 3x after deploy #847

Proactive

You'll hit memory limits in 3 days

Based on current growth rate

Cost

staging-worker is idle

4 requests in 7 days, costing $23/month

Automation With Visibility

Rules are pre-approved workflows. When conditions match, AI acts automatically. But you still see everything—because transparency doesn't stop when AI gets autonomy.

Create rules with natural language:
if cpu > 90% for 10min then scale up 2 machines
Got it. Rule "High CPU - Scale Up" is now active.

Scale Up — High Load

If job queue > 1,000 - Scale to 5 machines

12x this week

Scale Down — Low Traffic

If job queue < 3 - Scale to 1 machine

8x this week

Hot Potato

If CPU > 90% for 10min - Add 2 cores

3x this week

The Result

You sleep. Caveman doesn't.

But when you wake up, you can see exactly what happened. Every action. Every decision. Every reason.

AI that works with you, not for you.

AI you can trust—because you can see.