Product-minded full-stack work500K+ records processed at CDACMaps, computer vision, and adaptive interfacesFrom interface to deploymentProduct-minded full-stack work500K+ records processed at CDACMaps, computer vision, and adaptive interfacesFrom interface to deployment
FloodReady is a disaster-reporting and incident-mapping platform where authenticated users can submit geotagged flood evidence, see reports appear on a shared map, and receive background validation supported by image and weather analysis.
Led the project's end-to-end implementation across product design, frontend development, backend APIs, database architecture, validation services, containerization, and deployment.
Landing experienceLive product / 2026
Year
2026
Role
Team project · Primary developer / lead full-stack contributor
Flood information is fragmented, delayed, and difficult to evaluate at a local level.
02 / The approach
How it was shaped
Combine citizen-submitted geotagged evidence, shared mapping, and resilient background validation.
03 / Key decision
Where judgment mattered
Store and expose a report before background validation so a failed analysis cannot erase the original evidence.
Resilient report flow
Designed to preserve evidence
01
Citizen report
02
Authenticated API
03
Immediate persistence
04
Shared map visibility
05
Background validation
06
Severity and confidence update
02 / Human-centred product
EmoStudy
EmoStudy is an emotion-aware study timer that privately analyses facial expressions on the user's device and adapts Pomodoro sessions around their current state.
The product combines a focused timer experience with live expression scores, session insights, focus analytics, charts, and streak tracking.
Standard study timers ignore how the user is feeling during a session.
02 / The approach
How it was shaped
Use on-device expression analysis as a private interaction signal and adapt the timer experience around it.
03 / Product focus
Where judgment mattered
Make the experience useful even when camera permission is unavailable.
03 / Computer vision system
AquaVision
AquaVision is a lightweight computer-vision pipeline that restores degraded underwater images, improves visual clarity, and detects regions of interest without requiring expensive hardware or large training datasets.
One underwater frame, shown through restoration, enhancement, and the final detection output.
Structured data models, geospatial applications, containerized services, version control, and end-to-end deployment.
PostgreSQLPostGISMySQLPrismaDockerGitGitHubVercel
03 / Experience
LEARNING AT SCALE.
June 2024 - July 2024Centre for Development of Advanced Computing (CDAC), New Delhi
BIG DATA INTERN
Built a sentiment-analysis system that processed more than 500,000 records through Big Data ETL pipelines and presented the results through interactive visualization dashboards for stakeholder review.
Big Data ETLPythonNLPDashboarding500K+ records
Education / 2023 - 2027
B.Tech in Computer Science Engineering
K.R. Mangalam University, New Delhi
Involvement / 2024
Smart India Hackathon
University Team Member
Collaborated with a university team to prototype a software solution for a real-world problem under strict time constraints.
04 / About
I TURN MESSY, REAL-WORLD PROBLEMS INTO PRODUCTS PEOPLE CAN ACTUALLY USE.
I like turning messy real-world problems into products people can actually use. That means caring about the screen, the API, the database, and the deployment - not treating any of them as someone else's job. I learn by building, breaking, fixing, and eventually shipping.
Usually powered by curiosity, stubborn debugging, and an unreasonable number of browser tabs.
01 / See the whole system
The screen, API, data model, and deployment are one product—not separate hand-offs.
02 / Design before decoration
Sketching and conceptual drawing shaped how I think about layout, hierarchy, and the way an interface feels before a user reads a single line.
03 / Learn by shipping
Interested in full-stack products, useful automation, maps, and thoughtful UI.
05 / Contact
HAVE A PROBLEM / WORTH BUILDING?Let's turn it into something real.
Have an internship, graduate role, or product problem where a curious full-stack builder could help? Send the useful context directly.
Email is the clearest way to start. You can also find my work, profile, and résumé below.
