CUT TO:
INT. PROJECT ARCHIVE — STORYBOARD ROOM
The USER opens Dial4242 Operations/Analytics Dashboard.
DIVAKAR DESSAI
Let's run through the shots.
CASE FILE / Data Analytics / Full Stack / Geospatial
Dial4242 Operations/Analytics DashboardOperational Analytics & Demand Intelligence
Built a geospatial analytics dashboard for understanding historical ambulance demand. Processed raw Excel call data into office, district and state-level insights to support operational planning.
2025–2026
01–02 / OPENING SEQUENCE
Establishing the world
01 / Establishing Shot
The Problem
Operational ambulance data contained useful information about where and when demand occurred, but extracting meaningful patterns from raw historical trip records required extensive cleaning, aggregation and analysis.
02 / Wide Shot
The Context
The project was developed during a software engineering internship as an operational analytics platform for ambulance trip data. Users could upload historical Excel datasets and analyse activity across selected date ranges through a browser-based dashboard.
03 / CHARACTER NOTE
DIVAKAR'S ROLE
Subject
Divakar Dessai
Production
Dial4242 Operations/Analytics Dashboard
Take
03 / Role
Role notes
DIVAKAR DESSAI
I worked across the full stack, building the frontend dashboard, Python data-processing logic, Flask API endpoints, geospatial analytics and the integration between the analytics backend and the user interface.
04 / CLOSE-UP
THE ENGINEERING CHALLENGE
The core challenge was converting inconsistent operational trip data into information that could be explored quickly. This involved cleaning location and date information, resolving Indian pincodes, aggregating demand across multiple geographic levels and exposing those results through a usable dashboard.
05 / TRACKING SHOT
THE APPROACH
A plan emerges.
The system uses a Next.js and React frontend connected to a Flask backend. Uploaded Excel data is processed with Pandas, cleaned by date and location and aggregated into operational metrics. Pincode information is resolved into office, district and state groupings, while coordinate data is prepared for Mapbox-based geographic visualisation. Flask endpoints expose the processed analytics to the frontend.
06 / INSERT SHOTS
KEY FEATURES
The system takes shape.
07 / DIRECTOR'S NOTES
DESIGN DECISIONS
Things we decided along the way
01
Separated data-processing logic into a Python backend rather than performing heavy analytics in the browser.
02
Used Flask endpoints to expose processed analytics to the Next.js frontend.
03
Normalised six-digit Indian pincodes before performing geographic analysis.
04
Cached pincode lookups to avoid repeatedly resolving the same location.
05
Aggregated demand at multiple geographic levels so users could move between local and regional views.
06
Returned frontend-ready JSON structures from the backend to simplify dashboard rendering.
design decisions
somewhere mid-build
08 / RETAKES
WHAT WENT WRONG
Naturally, not everything cooperates.
09 / FINAL SHOT
THE OUTCOME
ProblemBuildOutcome
01
Delivered a full-stack ambulance analytics dashboard.
02
Created reusable backend endpoints for spatial and operational analysis.
03
Enabled ambulance activity to be explored geographically and over time.
04
Converted raw operational trip records into structured metrics suitable for decision-support dashboards.
10 / PRODUCTION NOTES
TECH STACK
The tools behind the scenes.
Next.jsReactPythonFlaskPandasMapbox GLRechartsXLSXTailwind CSS
11 / BEHIND THE SCENES
LINKS
FADE OUT.
USER closes the file.
One project down. A few more stories left.
DIVAKAR DESSAI
Pick another?