Products · AI-native infrastructure
A focused portfolio across language, execution, documents, and method.
We build a small, deliberate set of tools rather than a sprawling catalog. Each one is open-source-first where it can be, and every product carries an honest status label — Live, Beta, or Preparing — so you always know what you are adopting.
The portfolio
Four products, each solving a distinct layer of human–AI work. Status labels reflect real maturity, not aspiration.
EML
A high-density semantic-overlay language and transpiler for human–AI collaborative programming. It gives people and agents a shared, readable surface over the code that actually runs.
What it’s for: For teams that want humans and AI to read, write, and reason about the same program.
PHOSPHOR
Execution-as-Interface infrastructure that turns machine execution into a visible, operable surface for humans and AI agents. You watch a system run, then act on it directly.
What it’s for: For making opaque execution observable and controllable, instead of guessing from logs.
EveGlyph Editor
An AI-native, agent-native Markdown editor and document workspace. It is built for documents that humans and agents edit together, with structure agents can act on.
What it’s for: For writing, reviewing, and operating on documents alongside AI agents.
AMEP
Executable agent method packs: deterministic, auditable protocols that turn a methodology into something an agent runs, not a prompt it reads. Each pack ships schemas, a CLI, and a trace.
What it’s for: For turning a discipline into a repeatable, inspectable protocol an agent executes.
How they fit together
The four products are not a grab bag. EML is the language layer, PHOSPHOR the execution layer, EveGlyph the document layer, and AMEP the method and protocol layer — one coherent stack for human–AI work.
EML
A shared semantic surface so humans and agents work over the same program.
PHOSPHOR
Machine execution made visible and operable, not buried in logs.
EveGlyph Editor
Documents humans and agents author and act on together.
AMEP
Methodologies compiled into deterministic, auditable protocols.
Build with EVEMISS Technology.
Tell us what you are building, or read the source. Either way, you will find honest status labels and tools designed for humans and AI agents working side by side.