You are browsing as a guest. Sign up (or log in) to start making projects!

Jason_Mloza

@Jason_Mloza

Joined June 3rd, 2026

  • 3Devlogs
  • 4Projects
  • 1Ships
  • 15Votes
Builder, scientist, and AI developer. I love turning ideas into working projects and learning new technologies along the way.
Ship Changes requested

@Jason_Mloza on Alfred

What did you make?

I built Alfred, a personal AI assistant designed to run locally on your own machine. Alfred uses a FastAPI backend and local LLMs through Ollama to generate responses without depending entirely on cloud AI services.

The goal was to create an AI assistant that feels personal, customizable, and accessible while learning how modern AI systems work under the hood.

What was challenging?

The biggest challenges were connecting all the different parts together — building the backend API, integrating the local AI model, handling responses efficiently, and creating a smooth user experience.

Getting local models running on limited hardware was also challenging because performance depends heavily on available CPU/GPU resources.

What are you proud of?

I'm proud that I built Alfred from scratch and turned an idea into a working AI assistant. This project helped me learn more about APIs, AI models, backend development, and building complete applications.

Seeing Alfred respond and work as a real application has been one of my biggest milestones as a developer.

What should people know so they can test your project?

To test Alfred:

Clone the source code from GitHub.
Install the required dependencies.
Install Ollama and download a supported model.
Start the backend and frontend.
Open Alfred and start chatting.

Make sure your machine has enough resources to run a local LLM.

GitHub:
https://github.com/jasonmloza/Alfred

  • 3 devlogs
  • 2h
Try project → See source code →
Open comments for this post

@Jason_Mloza on Alfred

1h 4m 7s logged

Alfred is now alive. I built the React frontend, connected it to a FastAPI backend, and integrated Ollama for local AI inference. The assistant can now have conversations privately on my machine.

0
0
26
Open comments for this post
Reposted by @Jason_Mloza

@Jason_Mloza on Alfred

32m 42s logged

🧠 Alfred Devlog #2 — Giving My AI Assistant a Brain
Today I reached a major milestone with Alfred 🤖
When I started this project, the goal was to build my own AI assistant inspired by futuristic assistants like JARVIS — something that could understand conversations, help with tasks, and grow over time.
The biggest step this week was moving Alfred from an idea into a working AI system.
What I built:
✅ Built Alfred’s backend using FastAPI
✅ Integrated a local AI model using Ollama
✅ Created a /chat API endpoint for conversations
✅ Tested Alfred’s ability to receive messages and generate responses
✅ Cleaned up the GitHub repository and improved the project structure
Now Alfred can actually communicate:
User:

“Who are you?”

Alfred:

“I’m Alfred, your friendly and knowledgeable personal AI assistant…”

Seeing my own AI assistant respond through a system I built from scratch was honestly one of the coolest moments of this project.
How Alfred currently works:
User
↓
API Request
↓
FastAPI Backend
↓
Ollama AI Model
↓
Alfred Response
The next major step is building Alfred’s frontend interface — giving him a proper face and making the experience feel like a real assistant.
Future plans:

🎙️ Voice interaction

🧠 Memory system

🛠️ Tool integration

📱 Mobile companion app

⚡ More autonomous capabilities

This project is still evolving, but Alfred is officially alive. 🚀
#Stardance #HackClub #AI #Python

0
1
29
Open comments for this post

@Jason_Mloza on Alfred

32m 42s logged

🧠 Alfred Devlog #2 — Giving My AI Assistant a Brain
Today I reached a major milestone with Alfred 🤖
When I started this project, the goal was to build my own AI assistant inspired by futuristic assistants like JARVIS — something that could understand conversations, help with tasks, and grow over time.
The biggest step this week was moving Alfred from an idea into a working AI system.
What I built:
✅ Built Alfred’s backend using FastAPI
✅ Integrated a local AI model using Ollama
✅ Created a /chat API endpoint for conversations
✅ Tested Alfred’s ability to receive messages and generate responses
✅ Cleaned up the GitHub repository and improved the project structure
Now Alfred can actually communicate:
User:

“Who are you?”

Alfred:

“I’m Alfred, your friendly and knowledgeable personal AI assistant…”

Seeing my own AI assistant respond through a system I built from scratch was honestly one of the coolest moments of this project.
How Alfred currently works:
User
↓
API Request
↓
FastAPI Backend
↓
Ollama AI Model
↓
Alfred Response
The next major step is building Alfred’s frontend interface — giving him a proper face and making the experience feel like a real assistant.
Future plans:

🎙️ Voice interaction

🧠 Memory system

🛠️ Tool integration

📱 Mobile companion app

⚡ More autonomous capabilities

This project is still evolving, but Alfred is officially alive. 🚀
#Stardance #HackClub #AI #Python

0
1
29
Open comments for this post

@Jason_Mloza on Alfred

23m 30s logged

made good progress on Alfred, my AI assistant project.✅ Built the backend for Alfred.✅ Installed Ollama to run AI models locally.The next steps are:Install and connect a language model through Ollama.Integrate the model with Alfred’s backend.Polish the user interface.Build a mobile app so Alfred can be used on the go.My goal is to create an AI assistant that’s fast, private, and useful for everyday tasks. This is still an early version, but it’s exciting to see everything coming together.

1
0
29

Followers

Loading…