🧠 Project: LogiCopilot – Real-Time AI Copilot for Logistics
Track: Walmart Hackathon 2025 – Transforming retail supply chains

Tech Focus: RAG | Streaming | LLM | Anomaly Detection

🔍 Problem
Logistics decisions depend on fast-changing data — PDFs, invoices, safety rules, or shipment records. Delays in surfacing this info lead to unsafe dispatches, invoice fraud, policy violations, and access control breaches.

💡 Our Solution
LogiCopilot is an AI-powered copilot that continuously monitors incoming PDFs, CSVs, and webhooks using streaming + RAG + LLMs, to:

Flag driver safety risks instantly

Detect invoice non-compliance

Identify shipment anomalies or fraud

Enforce access policies in real-time

Users can ask natural questions like:

“Why can’t Driver Maya enter Warehouse 9?”
“Is Invoice 102 compliant?”
“Is Route 42 abnormal today?”
🧱 Tech Stack
Stream Processor: Pathway

Doc Ingestion: PathWay, pandas

LLM + RAG: LangChain + ChromaDB + GPT-4/Mistral

Anomaly Detection: Prophet, scikit-learn

Frontend: Streamlit and react js

Alerts: Twilio, Slack, Email

🚀 Impact
Faster decisions, safer operations, and full compliance across the supply chain—automatically and in real time.
#walmart
#sparkathon2025 #sparkathon #sparkathon2025#hackathon #walmarthackathon
#walmarthackathon
#sparkathon