"ALGOTITANS team on Vultr: Agent Rush Hackathon Hackathon"
Team Idea
We are building a **Safe CSV Analyst and Self-Healing AI Code Execution Agent** that allows users to upload CSV files, ask questions in natural language, and receive real results such as data analysis, charts, cleaned datasets, and execution reports. The agent will generate Python code using Vultr Serverless Inference, execute it inside isolated, resource-limited sandboxes on a Vultr VM, detect errors, and automatically attempt to fix and rerun failed executions. Each execution will have strict time and memory limits, restricted permissions, execution logs, and automatic cleanup to protect the main application and user data. We are looking for motivated team members with skills in Python, FastAPI, AI/LLM integration, Docker and sandbox security, frontend development, and cloud deployment. Our goal is to demonstrate a practical AI agent that executes real tasks while keeping failures and potentially unsafe code contained.
Team Members
.png&w=256&q=75)
Submission
Team Leader hasn't made a submission yet