agent m
software engineeringai agents
a dynamic and fully autonomous agentic ai trading portfolio companion designed to navigate the complex digital financial landscape. leveraging a multi-agent ai architecture, the platform transforms real-time market data, traditional news, and internet sentiment into personalised, actionable investment decisions.
features
- multi-agent ai architecture: sophisticated multi-agent system for autonomous investment decisions.
- real-time data ingestion: scrapes financial news and internet sentiment from yahoo, reddit, and x.
- nlp news analysis: uses spacy and finbert for event extraction and sentiment analysis.
- credibility & fact-checking: validates claims and assigns credibility scores to weighted sentiment.
- rag chatbot: natural language interface for querying news and portfolio context.
- autonomous trading agent: executes buy/sell orders via alpaca/ibkr based on market data.
- visualisation dashboard: real-time sentiment indicators, p&l trends, and holdings.
- notification system: alerts for breaking news and trade execution confirmations.
tech stack
FastAPI
Python
Go
LangChain
LangGraph
Next.js
Tailwind CSS
Qdrant
PostgreSQL
MongoDB
Redis
Docker
Terraform
Kubernetes
Kong Gateway
LLMs
Amazon Web Services