SEMICONDUCTORS — AI News Platform

Overview

SEMICONDUCTORS is an AI-powered analytics platform for semiconductor industry news. The system aggregates news from multiple sources, provides sentiment/volume charts, a RAG-based chatbot for research queries, daily summaries, and a searchable news feed with filters. Each news article has a dedicated page where clients can add tags and labels to improve model training.

Built for a single institutional client, this project evolved from initial foundations of the PUMP project but pivoted entirely toward news and research workflows.

My Role

Solo Full-Stack Developer — I built this project from zero to MVP, owning both backend and frontend development entirely.

• Designed and implemented the complete system architecture
• Built all frontend features in Nuxt 3
• Developed FastAPI backend services
• Created and maintained ETL pipelines in Prefect for news parsing and labeling
• Implemented token streaming for the chatbot using Langchain
• Collaborated with ML engineer on the RAG system — he focused on LLM prompts while I helped integrate Weaviate vector database with async engine support

Key Features

News Feed & Search

• Aggregated semiconductor news from multiple sources
• Filterable and searchable news feed
• Individual news pages with tagging UI for model training feedback

Analytics Charts

• News volume and sentiment trend visualization
• Time-series analytics tied to semiconductor themes and tickers

RAG Chatbot

• Research assistant powered by RAG (Retrieval-Augmented Generation)
• Weaviate as vector database for semantic search
• Real-time token streaming via Langchain
• Context-aware responses blending recent news with company knowledge

Daily Summaries

• LLM-generated daily market digests
• Structured briefings covering key themes, company events, and industry developments
• Designed as a "morning briefing" for the research team

Technical Implementation

Backend

• FastAPI for REST APIs (news, insights, chat endpoints)
• Weaviate vector database for RAG semantic search
• Langchain for LLM orchestration and token streaming

Frontend

• Nuxt 3 (SPA mode) with Vue 3 Composition API
• Vuetify component library
• Responsive layouts for charts, feeds, and chat interface

Data Pipeline

• Prefect for ETL orchestration
• Custom parsers for multiple news sources
• Automated flows for news ingestion, cleaning, LLM summarization, and tagging
• Scheduled jobs for reliable daily summary generation

Timeline

Tech Stack

Python

FastAPI

Prefect

ClickHouse

PostgreSQL

Nuxt

Vue

OpenAI

Azure

Redis

• Backend: Python, FastAPI, Prefect, Weaviate, Langchain
• Frontend: Nuxt 3, Vue 3, Vuetify
• Infrastructure: ClickHouse, PostgreSQL, Redis, Azure