dmonterocrespo

What would be the general purpose of storing the history in a remote database? Is it for use by agents? It's not the same as agents cloning the project and running "git log".

Hi HN!

I’ve been working on DaveLovable, an open-source experiment around AI agents for web development.

The core idea: Instead of a single LLM generating code, I use a small multi-agent system (Planner + Coder) that can:

* generate React apps * modify existing code * run terminal commands * commit changes to git automatically

Everything runs with a real Node.js environment in the browser using WebContainers, so the preview is not mocked.

I also added a visual editing mode (click on UI → modify styles) and multimodal input (you can upload mockups or PDFs and ask the agent to recreate them).

One interesting challenge was coordinating agents reliably without them getting stuck or over-planning.

Still early, but I’d love feedback, especially from people working on dev tools or AI agents.

Repo: https://github.com/davidmonterocrespo24/DaveLovable

Yes, the editor itself uses Monaco. That dependency is listed in the README

When I said "browser-based IDE", I meant the environment around it: project handling, compilation with arduino cli, the UI, the serial monitor, and the integration with the emulator and circuit simulation

Monaco is just the editing component

Thanks!

If anyone wants to try it quickly, there's a live demo available here:

https://velxio.dev/editor

If it also uses screen emulation, SPI protocols, IC2, and sensors, it's built from scratch. It also integrates the Arduino CLI for compiling the project and using external libraries. I created an algorithm for the wiring (currently working on it). I've also integrated more Arduino models and the Raspberry Pi Pico. I plan to integrate ESP32, but I haven't been able to finish the emulator using QEMU as a base. It's not just an avr8js frontend

If anyone wants a quick thing to try in the simulator, here's a simple LED blink example with Serial output:

void setup() { pinMode(13, OUTPUT); Serial.begin(9600); }

void loop() { digitalWrite(13, HIGH); Serial.println("LED ON"); delay(500);

  digitalWrite(13, LOW);
  Serial.println("LED OFF");
  delay(500);

}

In Velxio you can connect an LED to pin 13 and watch it blink, while the Serial Monitor prints the messages in the terminal.

Thanks! I'm glad you like it

Thanks! Hope you enjoy trying it out. feedback is always welcome.

Yes! Wokwi is great and definitely served as inspiration for Velxio. My goal with this project was mainly to explore how microcontroller emulators work internally and experiment with building parts of that stack myself.

Yeah exactly ,the Pico (RP2040) is a microcontroller rather than a full SBC, so it doesn't run a traditional OS like Linux.

In the future I'd like to support more low-level experimentation though (bare-metal programs, custom runtimes, etc.).

Thanks for checking it out!

I built specifically:

The browser-based IDE (editor, project handling, UI)

The circuit simulation layer that connects components to the emulator

The glue between the AVR8 emulator and the virtual peripherals (GPIO, UART, SPI, etc.)

The component interaction system (buttons, LEDs, displays, etc.)

The architecture that lets compiled Arduino sketches run and interact with the simulated hardware

Some parts like the AVR CPU emulation and the compiler toolchain obviously come from existing projects, but the goal of Velxio wasn't to re-implement an AVR core from scratch. It was to build a usable environment where all of these pieces work together in the browser.

I'm still having trouble connecting the cables and components properly. I'm looking for a better algorithm. I'm also trying to create a real-world electronics simulator in JavaScript using an engine like CircuitJS1.

Hi HN!

I built Velxio to explore how microcontroller emulators work internally.

It's an Arduino environment that runs entirely in the browser. You can write sketches, compile them with arduino-cli, and simulate circuits using a real AVR8 CPU emulator.

Features: - Arduino Uno / Nano support - Raspberry Pi Pico (RP2040) - GPIO, SPI, I2C, UART, ADC peripherals - interactive electronic components

The goal is to create an open-source environment for experimenting with embedded systems without installing anything.

I'd love feedback from embedded developers!

I built a small emulator project called Velxio.

A fully local, open-source Arduino emulator. Write Arduino code, compile it, and simulate it with real AVR8 CPU emulation and 48+ interactive electronic components,All running in your browser. GitHub: https://github.com/davidmonterocrespo24/velxio

The goal of this project was to learn more about how emulators work internally: CPU instructions, memory management, and low-level architecture.

It's still experimental, but it already runs basic instructions and I'm continuing to improve it.

An OutRun-style pseudo-3D racing game for the ESP32-S3 microcontroller with an ILI9341 320×240 SPI display. The same source compiles for both ESP32 hardware and Windows (via a Raylib-based emulator).

Features Pseudo-3D road rendering — segment-based back-to-front painter's algorithm, classic arcade style 3D player car — OBJ mesh (428 vertices, 312 triangles) with scanline affine texture mapping and shadow Traffic system — 6 AI cars with independent speeds and lane positions Procedural track — randomized curves, elevation changes, tunnels, and buildings generated at startup Dynamic scenery — pine trees, bushes, rocks, lamp posts, and 3D buildings with window patterns Day / sunset / night cycle — smooth color palette transitions every ~180 km of distance Atmospheric fog — exponential density toward the horizon Physics — acceleration, friction coast-down, centrifugal drift on curves, hill gravity effects Collision detection — crash with 2-second recovery Lap timing — current and best lap displayed on the HUD HUD — circular speedometer (0–300 km/h) with needle

https://github.com/davidmonterocrespo24/esp32s3-arcade-3d

DaveLovable is an open-source AI-powered web development platform inspired by Lovable.dev, v0 by Vercel, and Stitch by Google Labs. It combines cutting-edge AI orchestration with browser-based execution to deliver the most advanced open-source alternative for rapid frontend prototyping.

Unlike other tools, DaveLovable leverages:

Multi-agent AI orchestration (Microsoft AutoGen 0.7) with 40+ tools Google Gemini-3 Flash Preview (1M input tokens, 64K output) for blazing-fast code generation WebContainers (StackBlitz) for true Node.js runtime in the browser Visual editor mode with click-to-edit and drag-to-adjust controls Full Git version control with LLM-generated commit messages Real-time streaming of agent thought processes