Project index
Retail Simulation

ShelfSense

Assortment-planning workspace for testing SKU changes, delists, flowback, and scenario outcomes before shelf decisions go live.

Project
13 / 13
Role
Planning-system engineering
Stage
Private internal workspace
Primary stack
Streamlit + FastAPI
Project note

What the product covers.

ShelfSense gives category teams a planning workspace for assortment decisions. Users can review must-carry recommendations, edit SKU status or ACV, run simulations, save scenarios, export reports, and estimate flowback/margin impacts using Snowflake-backed data and model-driven simulation services.

Case journal

The decisions inside the build.

A concise account of the operating problem, the resulting system shape, and the technical decision that kept the work coherent.

01

Problem

Assortment decisions need a safe place to test delists, must-carry changes, ACV assumptions, and flowback effects before changes reach the shelf.

02

System shape

A Streamlit planning workspace connects Snowflake-backed assortment data with editable recommendations, saved scenarios, reports, and FastAPI simulation services.

03

Decision record

Scenario inputs and existing plan state stay visible together so planners can compare a proposed change with the latest known grid instead of editing in isolation.

Delivered scope

What the contribution actually covered.

This is a private internal-workspace summary. Company data, models, formulas, users, and performance outcomes are not disclosed.

  1. 01

    Assortment recommendation review and scenario editing surface.

  2. 02

    Simulation, saved-scenario, and report-export workflows.

  3. 03

    Integration between planning state, model outputs, and analytical data.

Build details

The working parts, without the theatre.

Technology
  • 01Streamlit
  • 02FastAPI
  • 03Snowflake
Publication boundary

This was a private engagement, so the public note is intentionally limited to the verified project scope.

This is a private internal-workspace summary. Company data, models, formulas, users, and performance outcomes are not disclosed.

No public destination published
Discuss a related project

Tell us what needs to be built—or made better.

Start an inquiry