Skip to main content

Systems that improve themselves, and the safeguards that keep them honest.

CTO · Applied AI · Miami Beach

I run a lab that invents AI systems, builds them, and keeps them running in production. The writing comes out of that work.

Anthusanth.us · Oct 2026

DeepSeek restrictions reach inside the tools

DHS's DeepSeek restrictions reach third-party tools, too. A separate poisoned-model experiment shows why knowing whose models are inside matters.

Read on anth.us ↗
Anthusanth.us · Oct 2026

Pi 1.0, and the Durable Agent We're Building on It

Earendil shipped Pi 1.0 and an experimental companion, Pi Durable. Pi is one of the agents Kanbus dispatches tasks to, and we like Pi Durable so much that we're building a durable Pi agent platform into Chatticus.

Anthusanth.us · Oct 2026

It's Not a System 1 Model If It Takes Ten Seconds

Fastino's GLiDE beats every open model on our reasoning benchmark, but not Jev: less accurate, five times slower, ten times the cost, and a confidence that barely tells right from wrong.

Anthusanth.us · Oct 2026

The OpenAI Decisions API Needs a Confidence You Can Trust

We've routed decisions on LLM confidence since 2023, and calibration is what we most hope OpenAI's new Decisions API gets right. The model underneath, GPT-6 Luna, isn't there yet: on 3,600 reasoning problems its 99% meant 68%, and by five…

Anthusanth.us · Oct 2026

Can Averaging Reduce Noise in an AI Decision Model?

We asked Jev the same question ten times in one request and expected ten identical answers. They split, four votes to six. Kahneman, Sibony and Sunstein call that kind of variation noise. Averaging the copies didn't make Jev more accurate,…

Piloboluspilobol.us · Sep 2026

The Next Chatbot Is Growing on This Year's Compost

Chatbots learned from everything people wrote in public. Now brands are writing fake posts in public for the chatbots to read, so the next chatbot is trained on them and repeats them as if they were true.

Piloboluspilobol.us · Sep 2026

The Button Marked Auto Has an Owner

Millions of programmers let a setting called Auto pick which AI answers them. Lately it keeps picking Grok, the one owned by Elon Musk, whose company also owns the program they are using, and who has shown little interest in checking which…

15M+automated QA decisions in production
180B+LLM tokens processed
$150M+ticketing revenue through a platform I built
30 yrsshipping software

Thirty years of inventing, building and operating software: a Perl webmail product when Hotmail was new, a BlackBerry app before the iPhone, a Las Vegas ticketing platform that cleared regulators and audits, and an AI lab that put a self-improving agent system into commercial production.

B.S. Computer Science, University of Florida. Engineering intern at 3Dfx during the original Voodoo Graphics chipset.

GitHubLinkedInanth.us/ryanRésumé (PDF)

2026 — nowDataParadeChief Technology Officer
2023 — nowAnthus AI SolutionsLead AI Solutions Architect
2024 — 2026Call Criteria / CapacityChief Scientist
2014 — 2024Tao Group / HakkasanVP Technology, VP Marketing Technology
2007 — 2013Venue DriverCTO
1997 — nowEndymion CorporationPresident