The problem
Inventory in fashion is unforgiving: order too much and cash is trapped in unsold stock; order too little and you miss the season. Pellemoda was managing this the way most SMBs do — spreadsheets, manual reorder points, and experienced staff making judgment calls under time pressure. The knowledge lived in people's heads, the data lived in disconnected places, and nobody had time to look ahead.
They didn't need a dashboard to look at the problem. They needed something that would act on it.
What I built
I built an autonomous inventory agent using LangChain multi-step, tool-use agents — not a single prompt, but an agent that reasons over the current state, decides what to check next, and takes action through real tools.
- Demand forecasting — the agent projects upcoming demand from historical patterns so reorder decisions are made against what's likely to happen, not what already did.
- Stock health monitoring — it continuously watches stock levels, flags items drifting toward stockout or overstock, and surfaces the why, not just the number.
- Supplier coordination pipelines — when action is needed, the agent drives the supplier-facing steps, turning "someone should email the supplier" into a tracked, automatic workflow.
Because it's built as a tool-use agent rather than a hard-coded script, it handles the messy, multi-step decisions that a rules engine can't — and its reasoning is observable, so the team can trust and audit what it does.
The results
- A reactive, manual inventory process became a proactive, autonomous one.
- Stock decisions are now driven by forecasts and continuous monitoring instead of spreadsheet reviews.
- The institutional knowledge that lived in staff heads is encoded in a system that runs every day without being asked.
Want the same?
If you have a repetitive, judgment-heavy process that a person does today — forecasting, monitoring, chasing, reconciling — there's a good chance an agent can own it. Book a 30-minute call and let's find out.