I Built My Dad a Local AI Translator for His Stock Research

Описание и применение

Приложение Argus Friend Brief переводит структурированные данные из панели Argus в простые, доказательно обоснованные обзоры, генерируемые локально на ноутбуке без передачи данных в облако.

Задача

My dad needed a way to digest Argus stock‑research data without navigating complex dashboards.

Результат

The app runs entirely on a laptop, keeps research data local, and produces evidence‑linked briefings that help the user understand each company’s situation.

Рабочий процесс

• Export a read‑only, sanitized snapshot from Argus SQLite
• Run local Gemma 3 4B via Ollama on the snapshot
• Validate JSON output against a Pydantic schema
• Check citations and safety guards
• Display briefing in Streamlit with expandable evidence panels

Доказательства

Argus Friend Brief takes a sanitized snapshot of that research and asks a local Gemma model to explain one company in plain language.

Источник

The model never sees the database or an open‑ended question. It receives one company's evidence catalog and a Pydantic‑derived JSON schema.

Источник

If a response contains an unknown citation or disallowed language, the app gives the model one repair attempt. If the repaired response still fails, the app displays an error instead of an unsupported brief.

Источник