Coding made easy. Build complex systems with basic commands and model interactions.
Sesi is a dedicated systems language designed for resilient stateful applications. No more boilerplate, just pure creativity and logic. Harness first-class reasoning primitives and distributed coordination natively.
let page_title = "Sesi"
let tagline = "Coding made easy. Build complex systems with basic commands and model interactions."
print "Generating Sesi landing page..."
# First-class reasoning primitive invoked natively
let html = model("gemini-3.1-pro-preview")
{"temperature": 0.2, "max_tokens": 8192}
{"Create a stunning landing page for Sesi."}
write_file("index.html", html)
print "Website successfully built!"
exec("start index.html")
Write less, do more. Sesi's syntax is stripped of boilerplate, allowing you to express complex logic and model interactions in plain text .sesi files.
LLM calls aren't SDK afterthoughts. They are native language primitives. Invoke models directly within your control flow just like standard functions.
Built for the modern network. Sesi natively handles distributed execution, allowing seamless coordination across multiple nodes without external message brokers.
Develop resilient stateful applications out of the box. Sesi automatically manages execution state, ensuring your long-running processes survive interruptions.
Powered by a robust tree-walking interpreter, Sesi evaluates Abstract Syntax Trees directly, providing predictable execution and straightforward debugging.
A true systems language for the AI era. Sesi provides low-level control over execution environments while maintaining high-level abstractions for reasoning.
Sesi operates on a unique execution model tailored for AI-driven logic. Unlike traditional compiled languages, Sesi utilizes a highly optimized tree-walking interpreter that evaluates .sesi files line-by-line. This allows for dynamic state injection and real-time reasoning adjustments.
| Phase | Description |
|---|---|
| Lexing & Parsing | Converts plain text .sesi files into an AST. |
| State Hydration | Loads resilient state from previous execution halts. |
| Tree-Walking | Interpreter traverses the AST, executing basic commands. |
| Reasoning Yield | Execution pauses natively when awaiting model inference. |
// Sesi Internal Tree-Walking Interpreter
private async evaluateExpression(expr: Expression): Promise<RuntimeValue> {
switch (expr.type) {
case 'ModelCallExpression':
// First-class reasoning primitive
return await this.evaluateModelCall(expr);
case 'Assignment':
return await this.evaluateAssignment(expr);
case 'StructuredOutputExpression':
// Natively handles AI structured outputs
return await this.evaluateStructuredOutput(expr);
}
}
# 1. Install the Sesi toolchain
git clone https://github.com/Misterscan/Sesi.git
cd Sesi
npm install
# Unlock the `sesi` command
npm install -g
# 2. Create a new Sesi file
touch hello.sesi
# 3. Write your first program
echo "print 'Hello, Sesi!'" > hello.sesi
# 4. Run the tree-walking interpreter
sesi hello.sesi
> Hello, Sesi!
Getting started with Sesi is incredibly straightforward. Because Sesi is a standalone programming language, there are no complex dependencies, no heavy SDKs to configure, and no boilerplate to write.
Your .sesi files contain everything needed to orchestrate complex, resilient stateful applications. Just write your logic, invoke your models, and let the interpreter handle the rest natively.