case study — academic project

Grazioso Animal Rescue

Rescue-operations dashboard · SNHU CS-340

A rescue-animal organization needed field data it could act on in real time. I built the CRUD backend and the dashboard that surface it.

Role: Backend Developer & Data Engineer
Stack: Python · MongoDB · PyMongo · Dash · Plotly · Pandas
Missions: Water · Mountain · Disaster

the problem

Grazioso Salvare — the course's fictional client — trains animals for emergency operations. The challenge wasn't collecting data; it was making it actionable in real time. Field staff needed to filter animals by rescue readiness, see deployment areas, and make data-backed calls fast, without hand-sifting a raw database.

what i built

MongoDB data layer

Schema and indexes designed for fast filtering by mission profile
Indexed queries for responsive field retrieval across the dataset
Structured to scale to the full shelter dataset without redesign

Python CRUD module

A reusable PyMongo module wrapping all create / read / update / delete operations
Input validation and explicit error handling — no bad writes, no silent failures
Consumed by a Dash dashboard used as a clean data surface, not decoration

Grazioso dashboard: filterable animal records table — fig 01 — mission-based filtering across the shelter dataset

Grazioso dashboard: outcome charts and map view — fig 02 — outcome charts + geolocation mapping

results

5,000+shelter records filtered

3rescue-role filters

CRUDfull create · read · update · delete

−80%manual record-sifting

A typical field query — water-rescue-suitable dogs under two years old ({"age": {"$lt": 2}, "mission": "water"}) — resolves to a filtered dashboard view with the map and charts updating together. Built against the course's Austin Animal Center dataset.

engineering notes

Resolved MongoDB ObjectId serialization for clean backend-to-frontend data transfer
Kept Dash lean — used purely as a data surface over the CRUD layer
Credentials isolated in a .env file, kept out of version control