Hello, I’m Evan Ma

Portrait of Evan Ma at the Xi'an Bell Tower

I like starting with real problems and turning ambiguous ideas into things that can run, be inspected, and keep improving.

I am exploring how people and AI can evolve together.

Today, I pursue that question through personal AI workflows, agent collaboration, task verification, handoff, memory, and a continuous record of how my judgment changes.

01 / Projects

Systems that begin with a question

View projects
01

Agent handoff protocol

SACP

When an agent claims completion, how can it leave a handoff another operator can inspect, judge, and continue?

Ongoing research
02

Inspectable execution record

Agent Flight Recorder

Can claims, evidence, gaps, and repairs from an agent run become an externally inspectable record?

Prototype validation
03

Web evidence agent

ShadowBuyer

Can inspectable evidence help hotels understand regional differences in prices, taxes, and terms?

Hackathon prototype

02 / Writing

How reflection becomes judgment

Read writing
TechnicalThe Essence of a Skill: Making Decision Processes ExplicitSkills as expert decision processes made explicit: deriving their design principles, boundaries, teaching uses, and business model from one signal-to-noise axiom.TechnicalHow to Build an AGENTS.md That Fits the Way You WorkA practical method for turning recurring failures into explicit, testable rules for collaboration, authorization, verification, and context.TechnicalFrom Tokens to Agent Cost Design: Reading AI Bills and Building Workflows That Do Not Burn ContextA first-principles note on tokens as cost, context, reasoning, output, and rework inside AI workflows.Build logVibe Coding Day 37: After BotLearn, I Don't Want to Remember Only First PlaceA build log about hackathon pressure, agent audit skills, and why the receipt layer became the center of the work.

03 / Research

Questions I am still trying to understand

View questions