Karan963
My journey at Kuanta

Kuanta Engine

Evaluating the transition from MVP to a scalable, Agentic AI architecture for startup analysis.

The Evolution Story

The MVP Monolith

Initial version was a tightly coupled application with synchronous processing. While functional, it struggled with timeout issues during long LLM analysis tasks and was difficult to extend with new features.

The Strategic Pivot

Recognizing the need for scalability, we architected a 'Modular Monolith'. This decoupled the analysis logic from the API layer and introduced asynchronous task queues to handle intensive AI workloads.

Agentic Architecture

We moved to an Agent-based approach. Specialized agents (Team, Finance, Market) now run independently, orchestrated by a central engine. This improved accuracy, error handling, and allowed for parallel processing.

Technical Innovation

Built to handle complex AI workloads with reliability and speed

Specialized AI Agents

A suite of focused agents (Team, Finance, Market, etc.) that specialize in analyzing specific aspects of a startup pitch deck.

Modular Design
Domain Expertise
Independent Scaling

Async Orchestration

Built on Celery and Redis to handle long-running AI tasks asynchronously, preventing timeouts and ensuring system stability.

Task Queues
Failure Recovery
Non-blocking API

Real-time Progress

WebSocket integration allows users to see detailed analysis progress in real-time as different agents complete their tasks.

WebSocket Events
Detailed Logs
Instant Feedback

Modular Monolith

A structured architectural approach that balances the simplicity of a monolith with the flexibility of microservices.

Clean Boundaries
Easy Testing
Simplified Deployment

Unified Data Layer

Seamless integration with existing databases (MongoDB) to fetch 'deck_metadata' and store structured agent results.

Schema Validation
No Data Duplication
Efficient Storage

Multi-LLM Support

Flexible AI core capable of switching between models (Gemini 2.0, OpenAI) for optimized performance and cost.

Model Agnostic
Cost Optimized
Gemini 2.0 Integration

Powered by Modern Stack

Python
FastAPI
Celery
Redis
MongoDB
Gemini AI
Docker