Your AI agent just did 9 things: pulled 412 orders, matched them to Stripe, caught a duplicate £248 charge, drafted a ledger entry, opened a ticket. Whole run cost $0.011.

Now the real design question — what does the user actually SEE?

A — The transcript. A paragraph of text and a collapsed "Used 3 tools".

B — The run canvas. Nine colour-coded rows: what ran, what it cost, what got flagged, and one step that stops and waits for you before money moves.

Same agent. Same run. Two completely different products.

Vote A or B — and tell me why. I build the B version for a living.

Wow! So beautiful and professional work!👌

Retail Inventory and Optimization Tool

Retailers often collect large amounts of sales and inventory data but lack a system to turn it into clear, actionable decisions, making it hard to know which products to reorder, which are slow-moving, and which risk stockouts or overstocking.

I built an end-to-end analytics pipeline to solve this. Using Python and Pandas, I cleaned a raw retail dataset, handling missing values based on what each one actually meant in context (e.g. a missing discount treated as "none," while a missing inventory value was filled with the average rather than falsely assumed to be zero). The cleaned data was loaded into MySQL for exploratory analysis, then connected to an interactive Power BI dashboard summarizing revenue, stock levels, demand fulfillment, and sales trends by category, store, and season.

Key findings included a 95.71% demand fulfillment rate (revealing real, recoverable lost sales), evenly balanced sales across all products categories and seasons, and one store carrying noticeably higher inventory than others despite similar sales flagging a possible overstocking risk.