🛠️ TrojanChat – Tech Stack This document outlines the technologies powering TrojanChat 2.0 AI.
Technology
Purpose
Next.js 14
React framework with App Router
TypeScript
Type-safe component development
CSS-in-JS (inline styles)
Scoped UI styling
Technology
Purpose
FastAPI
Async REST API + WebSocket support
Python 3.12
Runtime
Pydantic v2
Request/response schema validation
Uvicorn
ASGI server
Technology
Purpose
Groq
Ultra-low-latency LLM inference
Cohere
Text embeddings for semantic search
LangGraph
RAG workflow orchestration
Qdrant
Vector database for document retrieval
Technology
Purpose
Redis
Inference result caching (fail-open)
In-memory store
Chat message history (dev mode)
Technology
Purpose
Prometheus (prometheus-client)
API latency, cache hit/miss metrics
Structured logging
JSON logs via Python logging
Technology
Purpose
Docker
Containerized local and production deployments
Docker Compose
Multi-service orchestration (API + Qdrant + Redis)
GitHub Actions
CI/CD — lint, test, build on every push
Render
Cloud deployment target
Technology
Purpose
JWT / OAuth
User authentication
React Native
Cross-platform mobile app
Server-Sent Events
Streaming AI responses
PostgreSQL
Persistent chat history