A backend-first intelligent tool to upload CSV files and receive real-time insights, anomaly detection, missing data patterns, and schema suggestions — powered by Large Language Models (LLMs).
- Backend: FastAPI
- Language: Python
- AI Integration: LLM API (Ollama-Mistral & Gemini API)
- Planned Frontend: React (optional in later modules)
- Hosting: Local / Cloud-native (TBD)
This project was initiated as part of my personal learning and professional portfolio to:
- Build deep understanding of LLMs + data workflows
- Explore AI-assisted data debugging for real-world datasets
- Showcase startup-grade problem-solving for internships and part-time roles
| Feature | Description |
|---|---|
| ✅ FastAPI Setup | Basic backend server with health-check route |
| 🟡 CSV Upload + Schema Parsing | Extract structure, column types, nulls |
| 🔜 Anomaly Detection (LLM) | Detect outliers, missing logic, inconsistencies |
| 🔜 Natural Language Insights | Ask questions about your dataset |
| 🔜 Schema Suggestion Engine | Recommend schema/cleaning strategies |
| Module | Status |
|---|---|
| 1. Project Setup | ✅ Complete |
| 2. CSV Upload + Summary | ✅ Complete |
| 3. LLM-Powered Debugging | 🔜 In Progress |
| 4. Insights via Chat | ⏳ Planned |
Rupam Dusane
Aspiring AI x Data Infrastructure Engineer
LinkedIn
This project is in private development and not open source yet..
Please contact me directly for access, collaboration, or internship evaluation.