How I Built a Serverless AI Accounting App with AI assistant and Saved My Family from Spreadsheet Chaos
Link to the Source Code As a data engineer, I spend my days designing clean, optimized data structures. But at home, I face a much tougher crowd: my family. We manage our shared finances together to optimize our budget, and because of where we live and work, we have to do this in several different currencies (like USD, EUR, CZK, and UAH) Like any developer, I first tried to find a ready-made app…
This family created an accounting app using serverless AI technology to manage their finances in multiple currencies. Initially, they tried various existing apps but found them inadequate. They then decided to build their own solution from scratch. The main programming language used was Python, and they chose Django as their framework of choice.
A Star Schema was used for the database design, which proved to be an efficient structure for transaction bookkeeping. The app was divided into two main Django apps, members and transactions, and hosted locally on an Ubuntu server. In 2026, the family sought to improve the user experience by making the app serverless and adding a conversational AI assistant.
They built an assistant app within Django that could handle different intents like general chats, documentation, data retrieval, and commands. To ensure low-latency and cost-effectiveness, they utilized a hybrid pipeline with Groq and Azure OpenAI, optimizing the AI's responses with embeddings and vector databases. The entire process was streamlined by automatically ingesting document updates into the AI's knowledge base.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
