This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle
What I Built
PocketLedger AI is a lightweight expense organizer designed for freelancers, tutors, creators, and small businesses. It helps users record expenses, categorize transactions, review monthly spending, and export their records to CSV.
The application supports categorization through an open-weight model served by Ollama, with a transparent rules-based fallback when the model is unavailable.
Demo
https://pocketledger-ai-emv2vdb45nposflpkyt7gw.streamlit.app/
Code
https://github.com/godGamerz812/PocketLedger-AI
How I Built It
I built PocketLedger AI using Python and Streamlit, with Pandas for expense records and monthly summaries. The optional model integration uses Ollama to connect to an open-weight language model. The app also includes a rules-based fallback, CSV exports, automated tests, and a GitHub Actions workflow.
Why Does Open Innovation Matter?
Open-weight AI gives users greater control over model choice, deployment, privacy, and cost. By supporting Ollama, PocketLedger AI can use a model on infrastructure controlled by the user rather than requiring a closed, hosted AI API.
The open-source implementation allows developers to inspect, modify, and extend the project. Its transparent fallback also makes basic expense categorization usable when a model is not available.
My Agent Session
My First User
My first user is me. I built and tested PocketLedger AI to explore how a lightweight expense tracker can help organize daily spending, categorize expenses, and generate monthly summaries. My next step is to get feedback from freelancers or students who want a simple way to manage their expenses.
No paying customer yet — the goal is to validate the idea with real users and improve the app based on their feedback.
Prize Categories
Best Use of Gemma: PocketLedger AI supports optional expense categorization using the Gemma open-weight model through Ollama. The goal is to help freelancers and small businesses organize expenses while retaining greater control over where their data is processed.
Abhishek Kumar
Thanks for participating in Hacktoberfest 2026!
Top comments (0)