Researching Experiment 01
A personalised agent that feels human.
Research into agents that learn how one person works: they remember context, follow up and quietly keep things moving, the way a great human assistant would.
Rupesh Pasupuleti Sydney, Australia
Payments by day. AI agents by night. Notes from where the two meet.
Payments · AI · Research · Writing
Questions worth sitting with
01
What I believe
Payments will move by intent, not instruction.
You'll state the outcome you want. Software will choose the route, the rail and the moment.
The best payment is the one you never notice.
Friction was a tax we learned to accept. The next decade removes it from the moment of paying.
Regulation is a design input, not a constraint.
The teams that read the rules first build the products that last.
A few people with AI agents can now do what a department used to.
Leverage has moved from headcount to judgement.
Trust is the real infrastructure of money.
Rails move value. Trust decides whether anyone sends it.
02
Horizons
What I see coming
The Company That Runs While You Sleep
How a small company can be built from AI agents that work together on their own, and what the humans still need to own.
Read the essay · 5 minAn Agent That Feels Human
Today's agents compete on how much they can do. The one worth living with is defined by what it chooses not to do.
Read the essay · 5 minArchitecture Was Designed for Humans. It's Time to Revisit It.
Agent-built, human-accountable: the operating model for software after coding.
Read the essay · 4 minCoding Has Stopped. Understanding Hasn't.
As AI writes more of our software, the most valuable engineering skill is shifting from typing to understanding.
Read the essay · 4 minThe Wallet Learns to Think
Agentic payments won't arrive as a new rail. They'll arrive as a new kind of customer, one that never sees a checkout page.
Read the essay 2 October 2026 · 4 min read
The first customer who never sees a checkout page is already being built.
03
The Lab
Research & experiments
Agentic teams
How groups of specialised agents divide work, check each other and hand off cleanly.
Agents that pay
Experiments in how agents could hold mandates, spend within limits and leave receipts people can trust.