{"slug": "ledgerfriend-from-everyday-transactions-to-smarter-books", "title": "📒LedgerFriend — From Everyday Transactions to Smarter Books.", "summary": "A developer built LedgerFriend, an experimental bookkeeping assistant for Indian sole proprietors that uses a locally served Qwen2.5 model via Ollama to parse plain-language transaction descriptions into proposed accounting entries, which are then validated and posted using fixed rule-based logic with human approval. The prototype, a single-page HTML/CSS/JavaScript app persisting data in browser localStorage, deliberately keeps AI in an advisory role, with the developer noting that \"equal debits and credits are a necessary check—not proof that the accounts are correct.\" A user handover and feedback session with the intended sole-proprietor user is still pending.", "body_md": "*This is a submission for the [Hacktoberfest Weekend Challenge: Build for a Friend](https://dev.to/challenges/hacktoberfest-weekend-2026-10-01)*\n\nLedgerFriend is an experimental bookkeeping assistant for Indian sole proprietors, built for the Hacktoberfest 2026 Weekend Challenge: Build for a Friend.\n\nThe idea is simple: a small-business owner should be able to describe a transaction in ordinary language, review the proposed accounting treatment, and generate organised books without manually repeating the same information across multiple reports.\n\nAI suggests the transaction details. Fixed accounting rules generate the entries. A human approves what gets posted.\n\n⚠️ Prototype only. Not production accounting software, tax advice, or a replacement for a **Chartered Accountant**.\n\n👤 Who I’m Building For\n\nLedgerFriend is designed for an Indian sole proprietor who finds everyday bookkeeping difficult to organise and wants to understand how ordinary business transactions become accounting records.\n\nI have kept the person's identity and business details private rather than inventing a testimonial or publishing personal financial information.\n\nThe intended workflow is based on a common problem:\n\nRecording transactions in everyday language is easier than translating them into journals, ledgers, and financial statements.\n\nA user handover and detailed feedback session is still pending.\n\n**Please use fictional transactions only.**\n\nThe repository contains the **project source code and setup information**.\n\nThe intended user is an Indian sole proprietor who is a Friend, Who finds it difficult to keep everyday transactions organised and translate them into accounting records.\n\nI do not want to invent a user testimonial. The project becomes more useful when its design is based on one real person’s workflow.\n\nI started with a lightweight, single-page application using:\n\n| Component | Role | \n|---|---|\n| HTML | Application structure and forms | \n| CSS | Responsive dashboard and report styling | \n| JavaScript | Validation, posting rules, reports, and AI requests | \n| Browser localStorage | Prototype data persistence | \n| Ollama | Local model serving | \n| Qwen2.5 | Open-weight language model for transaction parsing | \n| VS Code | Editing and local development | \n| Python HTTP server / VS Code Live Server | Serving the interface locally | \n\nI installed and experimented with:\n\n`qwen2.5:3b`` qwen2.5:1.5b`\nThe smaller model was explored to reduce hardware demands. Smaller size does not guarantee adequate accounting classification accuracy.\n\nI used AI assistance to explore the idea, generate and revise code, and troubleshoot integration problems.\n\nTools used during the development process included:\n\nThe documented local inference path is Ollama + Qwen. I should only claim a Backboard runtime integration or a partner-category feature if it is present in the submitted implementation.\n\nThe intended flow is:\n\n**Description → AI suggestion → Validation → Human review → Rule-based entry → Reports**\n\nThe model proposes a supported transaction type and extracts fields such as amount, date, and reference.\n\nApplication code then selects accounts from fixed rules and calculates with integer paise. The owner must review and confirm before posting.\n\n★This does not eliminate mistakes. An incorrectly classified transaction can still balance.\n\n**Equal debits and credits are a necessary check—not proof that the accounts are correct.**\n\nUnclear notes belong in a review queue. That queue is not a suspense account, and it does not affect the books.\n\nThis prototype is intentionally narrow:\n\nI started with a simple question:\n\n**“Can AI create the accounts?”**\n\nBuilding LedgerFriend led me to a better one:\n\n**“How can AI make bookkeeping easier without silently becoming the authority?”**\n\nThat question became the foundation of LedgerFriend.\n\nAnd that is the project I want to keep building.\n\n**Thank you for taking the time to explore LedgerFriend!**\n\nThis project was a learning experience in building a **bookkeeping application, integrating local AI, debugging real-world problems, and separating AI suggestions from accounting rules**", "url": "https://wpnews.pro/news/ledgerfriend-from-everyday-transactions-to-smarter-books", "canonical_source": "https://dev.to/sushan_shetty_5ebec41a67d/ledgerfriend-from-everyday-transactions-to-smarter-books-353a", "published_at": "2026-10-03 11:56:52+00:00", "updated_at": "2026-10-03 12:08:32.306906+00:00", "lang": "en", "topics": ["ai-tools", "large-language-models", "ai-products"], "entities": ["LedgerFriend", "Ollama", "Qwen2.5", "Hacktoberfest", "VS Code"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/ledgerfriend-from-everyday-transactions-to-smarter-books", "markdown": "https://wpnews.pro/news/ledgerfriend-from-everyday-transactions-to-smarter-books.md", "text": "https://wpnews.pro/news/ledgerfriend-from-everyday-transactions-to-smarter-books.txt", "jsonld": "https://wpnews.pro/news/ledgerfriend-from-everyday-transactions-to-smarter-books.jsonld"}}