The Problem

Modern workflows involve repetitive manual tasks such as opening applications, creating files, and executing commands. These tasks consume time and reduce productivity. Additionally, non-technical users find it difficult to automate such processes without scripting knowledge.

Our Solution

We developed an AI-powered desktop automation assistant that converts natural language commands into structured actions using a Large Language Model (LLM). The system interprets user intent and executes system-level operations such as file creation and application launching using Python.

Solution Architecture

User Input → FastAPI Backend → Groq LLM (Intent Parsing) → JSON Output → Action Executor → System Command Execution → Response

Deliverables

– Working FastAPI-based backend API
– Groq LLM integration for intent detection
– Action execution module for system-level tasks
– End-to-end automation pipeline
– API testing via Swagger UI

Tech Stack

– Python
– FastAPI
– Groq LLM (LLaMA 3)
– Uvicorn
– OS & Subprocess libraries
– dotenv

Business Impact

This solution can significantly improve productivity by automating repetitive desktop tasks. It can be applied in enterprises for workflow automation, reducing manual effort and operational costs. It also enables non-technical users to interact with systems using natural language, improving accessibility and efficiency.

Project Snapshots (Minimum 5 Pictures)

1. Swagger UI Interface
2. API Execution Response
3. File Creation Output
4. Backend Code Structure
5. Command Execution Flow
Demo Video