E-commerce AI Recommendation Engine

I’m building a next-generation recommendation engine for my e-commerce store and need an AI specialist who can own the full pipeline—from data ingestion through to live deployment. The core goal is to serve shoppers with smart product recommendations, personalised offers and “frequently bought together” bundles that update in real time as behaviour shifts.

What I have now:
• A clean product catalogue (≈15K SKUs) with historical transaction logs and basic customer profiles
• A cloud environment ready for model training and API hosting

What I need you to deliver:
1. A scalable recommendation model (collaborative, content-based or hybrid—whichever best fits the data) that can output:
• Top-N product recommendations per user
• Dynamic personalised discount/offer suggestions
• Cross-sell bundles (“frequently bought together”)

2. REST or GraphQL endpoints so my dev team can query the engine from our storefront and marketing automation tools.

3. An evaluation report detailing precision/recall, lift over baseline and A/B test plan. Model retraining cadence and monitoring metrics should be included.

4. Handover assets: cleaned feature engineering scripts, model weights, deployment Dockerfile and concise README so future iterations are straightforward.

Tech stack is flexible; Python with TensorFlow, PyTorch or Scikit-learn is preferred, and AWS or GCP fits our current infra. If you’re comfortable with implicit feedback datasets, matrix factorisation, deep learning recommenders or reinforcement-learning approaches, that’s a plus.

Timeline is tight, so please outline your proposed approach, similar projects you’ve shipped and any questions about my data. Looking forward to collaborating on a recommendation engine that genuinely moves revenue.