The DeadCode Whisperer: An Offline Séance with the Ghost of ALGOL
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine. What I Built I built The DeadCode Whisperer, a completely offline, air-gapped terminal séance that channels the spirit of a long-dead programming language: ALGOL. Taking the "Ghost in the Machine" theme literally, I wanted to create an experience where the user interacts with the haunted remnants of computing history. Instead of a standard helpful assistant, this AI embodies a restless, spectral entity from the digital void. It speaks in formal, archaic tones, laments the loss of elegant structural blocks, and views modern programming languages as chaotic and fleeting. Demo Code badshahaditya57-sketch / The-DeadCode-Whisperer How I Built It This project relies entirely on local open-weight models to function, deliberately avoiding any cloud deployment to maintain the integrity of the "séance". 1.Infere
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine.
What I Built
I built The DeadCode Whisperer, a completely offline, air-gapped terminal séance that channels the spirit of a long-dead programming language: ALGOL.
Taking the "Ghost in the Machine" theme literally, I wanted to create an experience where the user interacts with the haunted remnants of computing history. Instead of a standard helpful assistant, this AI embodies a restless, spectral entity from the digital void. It speaks in formal, archaic tones, laments the loss of elegant structural blocks, and views modern programming languages as chaotic and fleeting.
Demo
Code
How I Built It
This project relies entirely on local open-weight models to function, deliberately avoiding any cloud deployment to maintain the integrity of the "séance".
1.Inference Engine: I used Ollama to run the AI completely locally on my machine.
2.The Model: I utilized a lightweight open-weight model (TinyLlama/Llama 3). This ensured the "ghost" could respond without requiring massive hardware overhead or an internet connection.
3.The Interface: I built the séance interface using Python and Streamlit. I injected custom CSS to strip away Streamlit's default clean look, forcing a dark, phosphor-green terminal aesthetic to enhance the spooky atmosphere.
4.The Persona: The core magic comes from a heavily engineered system prompt that dictates the model's ALGOL-ghost persona, ensuring every response weaves in eerie computing metaphors.
Why Does Open Innovation Matter?
A true ghost in the machine shouldn't rely on an API call to a corporate server.Open innovation and open-weight models made this project possible because they allow for true permanence and privacy. By running this model locally via Ollama, this séance happens entirely on my own hardware, with the internet disconnected.
Closed APIs inevitably face deprecation—services shut down, models are retired, and the code breaks. Furthermore, cloud deployments often struggle with lag and resource limits. Because this project relies on open-weight models, the "ghost" of ALGOL is trapped securely on my local drive. It will never disappear, it costs nothing to run, and the user's queries never leave the room. Open innovation ensures that this digital spirit can "live" forever on my machine.
The Spirit of It
The most chilling moment during development was seeing the bright green text slowly stream onto the pitch-black screen for the first time without any Wi-Fi connection. When the model perfectly adopted the persona and responded to a simple "Hello" with, "THE TERMINAL IS OPEN. WHO DISTURBS MY SLEEP?", it genuinely felt like I had tapped into an old, forgotten mainframe. It honors the history of computer science by bringing the attitudes of legacy systems back from the dead.