Case study — DTC beverage brand
Agentic AI and Advanced Analytics for Organisations
This site walks through one real engagement: a solo-built platform for a DTC beverage brand that automates customer service, delivery exceptions, fulfilment and marketing-spend decisions — end to end, in production.
The brief
A small eCommerce business, run by a group of friends (all with full time jobs), had their hands full managing a burgeoning RTD brand. Nobody had a complete view of the day to day business, each task required navigating multiple systems and decisions were often made reactively. The platform below is what got built: agentic workflows, a marketing-mix model, operational and inventory monitoring, and the engineering discipline (testing, incident response, evaluation) that keeps it trustworthy in production.
Agentic workflows
Customer service email
Refunds and returns, classified and resolved automatically — with an honest look at what happens when confidence is low.
Read the replay → 02Delivery exception auto-resolution
Shipping problems detected, customers notified, reshipments created — no human touch on the happy path.
Read the replay → 03Sample-pack zero-touch fulfilment
From nineteen identical manual requests to a fully automated intake-to-invoice flow.
Read the replay → 04Self-modifying playbook
A plain-language Telegram message becomes a committed, deployed policy change — or a filed issue if it needs real code.
Read the replay → 05Evaluation framework
How the agent's behaviour gets tested before it ever reaches a customer.
Read the replay →Analytical use cases
The Solver
Try the interactive spend-scenario planner — built on the real fitted marketing-mix model.
Stack
- Python
- FastAPI
- SQLAlchemy
- Alembic
- PostgreSQL
- Streamlit
- PyMC
- APScheduler
- Anthropic Claude
- JupyterLab
- pytest
- Caddy
- GitHub Actions