the notebook of

Jamal Lyons

Computer Science Student @ Georgia State University

Hi, I'm Jamal. I love building the invisible layers that make software feel alive: real-time servers, APIs, and systems that are fast, reliable, and built to scale.

This notebook is where I keep that work. Inside you'll find:

  • backend architecture
  • systems programming
  • open source
clientthe part I likedata

the notebook of

Jamal Lyons

Computer Science Student @ Georgia State University

Hi, I'm Jamal. I love building the invisible layers that make software feel alive: real-time servers, APIs, and systems that are fast, reliable, and built to scale.

This notebook is where I keep that work. Inside you'll find:

  • backend architecture
  • systems programming
  • open source
clientthe part I likedata

the notebook of

Jamal Lyons

Computer Science Student @ Georgia State University

Hi, I'm Jamal. I love building the invisible layers that make software feel alive: real-time servers, APIs, and systems that are fast, reliable, and built to scale.

This notebook is where I keep that work. Inside you'll find:

  • backend architecture
  • systems programming
  • open source
clientthe part I likedata

About

I'm a Computer Science student at Georgia State University with a growing passion for backend architecture and systems programming.

I'm constantly exploring how things work under the hood, whether that's designing efficient APIs, optimizing data flow, or diving into low-level code to understand what really makes computers tick.

For me, development isn't just about writing code. It's about crafting systems that are fast, reliable, and built to scale.

the app you seeAPIsservers + networksmemory, CPUhere
where I like to spend my time

Problems I like

The ones that sit below the interface, where speed and reliability get decided.

  • Real-time servers. Keeping many connected clients in sync, and notification systems that reach people the moment something happens.
  • APIs. Designing interfaces that are efficient and hard to misuse.
  • Data flow. Finding where time goes between a request and a response, and cutting it.
  • Low-level code. Networking, memory, and performance. Understanding what really makes computers tick.
  • Developer tools. Type safety, debuggers, and setup scripts that remove whole kinds of bugs.
beforeafter← faster

Projects

  • A Rust-native debugger built from the ground up to understand Rust's ownership, lifetimes, and memory model.

  • A statically typed systems language that compiles to portable bytecode and runs on the Phoenix VM.

  • A calm, local-first desktop music player for the music you own. No streaming, no subscriptions. Library, playlists, tagging, and a 10-band EQ.

  • End-to-end type safety for Convex in Rust. Generates Rust types straight from your Convex schema.

  • Set up a Mac for coding from one repo: Homebrew apps, Fish shell, and editor settings, with a preview of every change before it happens.

  • A shared collection of Agent Skills for Claude Code, Codex, Cursor, and Gemini CLI. Edit once, use across agents.

  • A lightweight, zero-dependency HTTP client for type-safe API requests in JavaScript/TypeScript.

and more on GitHub ↗

Ferros

a Rust-native debugger · github.com/JamalLyons/ferros ↗

6 fn main() {7 let seed = black_box(40);89 let result = add_one(seed);10 println!("result: {result}");11 }breakpoint: the process is pausedhere, before this line runs
step goes into add_one, next runs it and stops on line 10

Most debuggers were built for C and learned Rust later. Ferros is built from the ground up for Rust's ownership, lifetimes, and memory model.

It runs on macOS on Apple Silicon today. It can launch or attach to a process, set breakpoints, single-step, and read registers and memory. The core engine has zero unsafe code and no OS dependencies, so other platforms can plug in later.

Convex Typegen

Convex schema → Rust types · github.com/JamalLyons/convex-typegen ↗

convex/schema.tsnotes: defineTable({
 text: v.string(),
 priority: v.nullable(
   v.union(v.literal("low"),
     v.literal("normal"),
     v.literal("high"))),
 tag: v.optional(v.string()),
 done: v.boolean(),
})
generated Rustpub struct NotesDoc {
 pub _id: Id<tables::Notes>,
 pub done: bool,
 pub priority:
   Option<NotesDocPriority>,
 pub tag: Option<String>,
 pub text: String,
 …
}
  • v.nullable → Option; the three literals become an enum
  • v.optional → Option too: the key may be missing
Each schema field maps to a typed Rust field: text to String, priority to an optional enum, tag to an optional String, done to bool.

Convex gives TypeScript developers end-to-end types for free. I wanted the same in Rust. Convex Typegen reads your schema and function signatures at build time and turns them into Rust types.

A wrong table, field, argument, or return type becomes a compile error instead of a runtime bug.

dotfile-sync

a Mac, set up from one repo · github.com/JamalLyons/dotfile-sync ↗

fresh Mac> fishready_plan
$ dotfilesync plan
Plan p-… (profile base)
[low] [irrev] [brew] brew install fish
[medium] [irrev] [brew] brew install --cask zed
[medium] [rev] [copy] copy ~/dotfiles/fish/config.fish -> ~/.config/fish/config.fish
[high] [rev] [shell] set login shell to /opt/homebrew/bin/fish

4 actions: low 1, medium 2, high 1; reversible 2, irreversible 2

nothing has changed yet. apply asks first, then Touch ID

dotfile-sync installs my Homebrew apps, Fish shell, and editor settings from one repo. It compares the Mac with the list, shows exactly what it would do, and only acts after I say yes. Every file it replaces is backed up first.

Writing

My reading list of things I've written.

  1. · 15 min read

    Transactional Email in Convex with Bluefox and AWS SES

    Bluefox is a modern email API built on top of Amazon SES, designed for developers and SaaS teams. Whether you're sending transactional emails or integrating real-time webhooks in a serverless Convex backend, Bluefox simplifies the process.

  2. · 25 min read

    Authentication Best Practices: Convex, Clerk and Next.js

    Authentication is the backbone of any full-stack application, but it's also one of the easiest places to introduce subtle, hard-to-debug security flaws, especially when scaling an app to a global audience.

  3. · 15 min read

    Building Type-Safe Rust Applications with Convex

    Convex offers a unique combination of developer experience, serverless functions, and real-time subscriptions, and under the hood it's powered by Rust. Here's how to get end-to-end type safety from Rust with Convex Typegen.