01Retail
Inventory copilot for a 30-store chain
Replaced manual stock checks with a forecasting agent that cut out-of-stocks by 38% in one quarter.
−38%stock-outs
Client
Regional grocery & home essentials chain · 30 stores · Tier-2 India
Engagement
10 weeks · Premium engagement
Stack
Python · Time-series forecasting · GPT-4o · PostgreSQL · Slack
01The Challenge
What we walked into
Store managers were running stock checks manually each morning and ordering by gut feel. Out-of-stocks on fast-movers were costing the chain roughly 7% of weekly revenue, and over-orders were tying up working capital in the wrong SKUs.
02The Approach
How we tackled it
- 01Pulled 18 months of POS data and built a per-SKU forecasting model that accounts for weekday, weather, and local events.
- 02Wired the forecast into a daily Slack digest that ranks the top 50 reorder candidates per store with one-tap approval.
- 03Trained store managers in two on-site workshops and shadowed three of them for a week to tighten the digest format.
03The Outcome
What changed
- 38% reduction in out-of-stock SKUs across 30 stores within one quarter.
- Working capital tied up in slow-movers fell 22%.
- Manager time spent on ordering dropped from ~45 min/day to under 8 min/day.
“
We expected the model to be right sometimes. We didn't expect it to be right often enough that our managers stopped second-guessing it. That's when we knew it had landed.
Have a similar problem?
Let's build something worth using.
Tell us about the bottleneck. We'll come back with a free 30-minute call and a quick read on what's feasible.
Start a conversation→More work
View all →Healthcare
Triage assistant for an outpatient network
Voice + chat intake that routes patients in under 90 seconds and books the right specialist on the first try.
90s
avg. triage
Logistics
Document automation for cross-border shipping
OCR + LLM pipeline that ingests bills of lading, flags discrepancies and clears paperwork without an analyst in the loop.
6×
faster clearance