TYMOFII
PIDLISNYI
Hover here
An open protocol for what an AI agent is allowed to do, and proof of what it did.
It checks authority before execution and leaves a signed record after. Open source, with SDKs in four languages and an IETF Internet-Draft.
The gateway built on APS.
The protocol says what governance is. The gateway says how well it works: it sits between an agent and the systems it can touch, decides, and records the outcome. APS stays open. The gateway is where the business gets tested.
My attempt to make social media disappear as a job.
You talk, the system turns it into content, and agents handle the rest. We reached 300+ users and 1.6M monthly impressions.
I started ofNature after Starbucks told us our first sustainable straw could not scale.
We built a cellulose acetate alternative, raised $150K, and set up manufacturing in Ukraine. War ended the company.
In 2025, McDonald's adopted the same material globally.
Things I build because I want to know what happens.
An experiment in meeting people through AI instead of feeds.
Tell your agent who you want to meet. It finds matches. Both people opt in before anything happens.
120+ profiles are live.
Explore βSmall apps and weird ideas I build fast to see if they deserve more time.
Some become products. Some are just a weekend well spent.
Explore βA prototype built around one question:
If AI companies make money from your data, should some of that value come back to you?
You choose what to license. If it sells, you get paid.
Learn More βI wanted to make breathing measurable, so I built a nose wearable with airflow sensors and an AI coach.
The hardware worked. I paused it when the next step clearly needed a medical cofounder, not more solo building.
Full Project βA simulation of 67 agents monitoring one patient at once.
The question was whether agents updating each other's beliefs could surface risks that disappear when you simply average their answers.
Watch Demo βI use AI to make music because some stories are easier to sing than explain.
Ten albums in two months. A lot of them are about love that never happened.
Listen βBefore startups, I directed films and music videos. Some reached 14M+ views.
I still come back to visual storytelling whenever code is the wrong medium.
View All βPublished papers, drafts, and ongoing work on AI agent governance.
Defines the three-layer protocol β Agent Passport, Human Values Floor, Beneficiary Attribution β that grounds the Agent Passport System.
Authority attenuation formalized β delegation can only narrow, never expand. Formal invariants and adversarial testing.
Formal model for how agent authority narrows monotonically across delegation chains. Independently validated by external research groups.
Lifecycle controls for derivative AI artifacts β telemetry scopes, behavioral measurement objects, bring-your-own-model boundaries.
Binding physical-layer attestation to delegation chains. Seven experiments on IBM Quantum showing measurable Bell and GHZ fidelity gaps.
The unit of agent governance is the population-with-medium, not the individual agent. Six rounds of adversarial review.
Cryptographic commitment over which sparse-autoencoder features were active during a decision. Envelope spec, three-stage verification, four dispute primitives. Experimental results against Llama-3.1-8B via Neuronpedia.
Cryptographic governance proves procedural validity, not effect safety. Five omitted-variable classes catalogued, defeat traces in reference implementation.
A controlled isolation of the retention rule in populations of short-lived LLM agents. Peer-voted retention kept a worse shared memory than random selection in 4 of 5 replicates. Pre-registrations, kill conditions, and artifact record included.
IETF Internet-Draft formalizing the Agent Passport System interoperability surface for cross-protocol agent identity, delegation, and enforcement.