AI Inventory & Support Optimization

My logistics operation needs a practical AI layer that tightens our inventory management while simultaneously powering automated customer support.

Right now stock levels are tracked manually in spreadsheets and routine shipment questions funnel through a small human team. The goal is a single, integrated solution that:

• Forecasts demand and safety stock in real time using historical order data, seasonality and transit times.
• Flags slow-moving or at-risk SKUs and recommends redistribution before they tie up capital.
• Feeds those insights directly into an AI-driven chatbot/voice bot able to answer “Where’s my order?” and similar status queries without human intervention, pulling ETA and stock data from our WMS/TMS API.

Tech environment: Shopify storefronts, a MySQL warehouse database, ShipStation for fulfillment and Slack for internal alerts. Python, TensorFlow, or similar open-source frameworks are welcome; I am open to SaaS APIs if they meet our data-privacy standards.

Acceptance criteria
1. Inventory forecast MAE below 5 % for top-moving SKUs over a 30-day test window.
2. Chatbot resolves at least 70 % of tracking inquiries unaided during the pilot week.
3. Clean, well-documented codebase with setup instructions so my in-house dev can maintain it.

In your proposal focus on your experience creating or deploying AI models for logistics, supply chain, or e-commerce—you can skip lengthy boilerplate. A concise portfolio link or brief case study that proves you have done this before will carry the most weight.