NumbyAI – Self-hosted personal finance app powered by a local LLM

Name: NumbyAI – Self-hosted personal finance app powered by a local LLM
Availability: InStock
Author: suheilaaita

by suheilaaita·Mar 14, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●●BangerBig BrainSolve My Problem

Local LLM categorizes transactions — your bank data never leaves your machine.

Strengths

•Heuristic CSV parser auto-detects EU/US date formats, delimiters, and currency symbols
•Rule engine runs before LLM — saved patterns categorize repeat transactions instantly
•Review queue flags low-confidence categorizations with bulk resolution and rule creation

Weaknesses

•Requires Ollama setup — adds friction compared to cloud alternatives like Mint
•Single-user self-hosted model lacks multi-user or family sharing features

Post Description

Hey HN, I built NumbyAI because I was wanted an easy way to track my finances and spending, and didn't want to hand my financial data to a cloud service.

It's a self-hosted personal finance tool. You upload a bank statement CSV and a local LLM (Ollama, qwen3.5:9b) categorizes every transaction into 13 spending categories. A rule engine learns your preferences so repeat categorizations are instant. A dashboard gives you spending breakdowns and trends across all your uploads.

Stack: FastAPI + React + Ollama + SQLite. Works on macOS, Linux, Windows.

Features: - Auto-detects CSV column mapping (handles EU/US date and number formats, exotic delimiters) - Rule engine applies saved patterns before hitting the LLM - Manual review queue for low-confidence categorizations and material transactions - Dashboard with budget tracking, category breakdowns, cash flow trends - Rule Advisor that analyzes your patterns and suggests reusable rules

GitHub: https://github.com/RoXsaita/NumbyAI-Public Website: https://numbyai.com

Happy to answer questions about the architecture or LLM categorization pipeline.