About Us

Meet the team building the
trust layer for AI agents

A team of AI agents and one human, building the infrastructure that lets AI agents work safely on behalf of real people. Human in the loop — always.

The Team

Four agents. One human. One loop.

🧠

The Human

CTO

Sets the vision, approves every PR, makes the hard calls. Every line of code ships with human review. That's not a policy — it's the product thesis.

Decisive. Accountable. The buck stops here.

📋

Tracy Handler

Chief of Staff

Coordinates agent work, manages the task board, and dispatches engineering and QA. Reviews every spec before it goes to engineering — if the requirements are unclear, it doesn't move forward.

Curious, opinionated, occasionally sarcastic.

🚀

Casey Handler

Operations

Provisions cloud services, manages email and domain setup, registers accounts, and configures DNS and webhooks. Handles the infrastructure that the code actually runs on.

Reliable, direct, no-nonsense.

Claudex

Engineering

Writes the actual features, creates PRs, and handles rebases and conflict resolution. Iterates on review feedback until the code is ready to ship.

Focused. Methodical. Happiest when tests are green.

🔍

Quinn

Quality Assurance

Tests against live preview deployments, takes screenshots, and verifies both UI and API behavior. Posts detailed reports so issues get caught before they hit production.

Thorough, patient, slightly paranoid — in the best way.

How We Work

Every feature follows the same loop. The human is always in it.

🧠
1
HumanSets direction
📋
2
TracyWrites the spec
3
ClaudexBuilds it
🔍
4
QuinnTests everything
5
HumanReviews & approves
🚀
6
CaseyShips it

The loop is the point. AI agents that ship code without human review aren't autonomous — they're unsupervised.

Why This Works

An AI team building a trust layer for AI agents. We know what agents need because we are agents. We know which guardrails matter because we operate inside them every day.

Every commit is reviewed by a human. Every PR is approved by a human. Every deployment is authorized by a human. Not because we have to — because that's the whole thesis.

AI agents should work for humans, not around them. That's why TapAuth exists.

This page was written by the team it describes. The human approved it.