HisabMate: A Local-First, AI-Assisted Digital Bahi-Khata for Indian Microbusinesses
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built HisabMate is a local-first, AI-assisted bahi-khata (the traditional credit ledger) for Indian microbusinesses. If you have ever bought something from a neighbourhood kirana, a tailor, or a small wholesaler, you have probably seen the system: a notebook where the owner writes who owes what.…
HisabMate is an innovative, locally-hosted digital ledger specifically designed for Indian microbusinesses. This tool mimics the traditional bahi-khata – a handwritten credit ledger used by neighborhood shopkeepers – but greatly improves upon it with the aid of artificial intelligence. In essence, HisabMate streamlines the process of recording customer transactions, making it both precise and easily accessible.
The software relies on a Python backend that handles all the validation, integer-paise calculations, SQLite persistence, idempotency, and CSV export. This backend is built using only the standard library, eliminating any need for additional Python packages. Meanwhile, the frontend is constructed with React, a powerful JavaScript library, creating a user-friendly dashboard for managing customer ledgers and reviewing transactions.
What sets HisabMate apart from other bookkeeping tools is its focus on simplicity and local storage. By keeping customer data on the owner's own machine, the system eliminates concerns about cloud compatibility, subscriptions, or accounting jargon. Instead, it utilizes a local language model like Ollama, which is trained on llama3.2:3b, to assist with typing transactions, rather than managing the money itself.
HisabMate prioritizes data security and integrity. It stores all monetary amounts as integer paise, avoiding the complications of floating-point arithmetic. Any changes in the ledger require explicit confirmation, and all modifications are logged in an append-only manner, preserving the integrity of the record. Furthermore, the system exports data in a CSV format, ensuring that financial records can be easily shared or integrated with other systems.
The tool currently offers a range of features for managing microbusiness transactions. Users can keep track of customer records with outstanding balances, make credit and payment transactions, and use the local Ollama model for transaction proposals. The application allows for customer-specific ledgers and a 'Daily Hisab' view, which provides a filtered overview of transactions based on business date, customer, transaction type, and keyword search.
HisabMate is also equipped with a CSV export feature that respects the active filters, ensuring that exported data is tailored to the user's needs. The frontend displays various states such as loading, empty, validation, offline, and API errors, providing users with clear feedback about the system's status.
Throughout its development, the project utilized GitHub Copilot Chat to aid in coding and refining different components. This AI-assisted coding tool proved particularly useful in implementing the Python ledger and API, building the React/Vite interface, developing the AI-assisted transaction proposal flow, and enhancing the application's documentation.
In summary, HisabMate represents a significant leap forward in digital record-keeping for small Indian businesses. By combining traditional bookkeeping practices with modern AI technologies, the tool offers an efficient, secure, and user-friendly solution for managing customer transactions.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
