Lokendra Singh.
← ALL PROJECTS

GovTech · Rural Chhattisgarh · CGSRLM & Pradan

Samriddhi

Offline-first data platform: field workers capture data with no signal, and it syncs with conflict resolution into a central analytics layer.

LEAD ENGINEERAndroidDjangoDruidSuperset

1,000+self-help groups
50,000+beneficiaries
40%faster processing
Offlinefirst, by design
Client
PRADAN with CGSRLM, Chhattisgarh
Domain
Rural livelihoods, GovTech
My role
Offline-first app and data platform
01

Context

  • Field workers visit villages with little or no signal
  • Data had to be checked before anyone used it
  • Programme teams needed district and block-level answers
02

Before and after

Data collectionPaper and phone calls, often re-enteredOffline app with automatic sync
Data qualityUnchecked, inconsistent formatsFixed formats and supervisor approval before use
VisibilityNo central pictureDistrict and block dashboards
Processing timeManual consolidationDown 40%
ReachProgrammes designed on partial dataBenefit distribution to 50,000+ beneficiaries guided by data
03

The problem

  1. No signal in the fieldData collectors work in places with weak or no internet, so a web form simply could not be the tool.
  2. Data you can't trust yetField data has to be complete, consistent and in a fixed format before anyone decides anything with it, and supervisors needed a way to check it.
  3. Numbers without meaningRaw survey answers did not tell a programme manager which district or block needed what.
  4. LanguageMost field workers are more comfortable in Hindi than English.
04

How it works

05

What we built

  1. Works offline
    • Manage groups and members
    • Collect and save data with no internet
    • Syncs on its own when signal returns
  2. Safe sync
    • Same record can change in two places
    • Conflicts resolved explicitly
    • No field worker's data is ever lost
  3. Supervisor check
    • Data arrives as unvalidated
    • Reviewed and approved before it counts
    • Every step logged
  4. Clear dashboards
    • Overall picture down to district and block
    • In Hindi and English
05

Key insight

Design for the worst network your users have. In rural India, offline isn't an edge case, it is the main case.
06

The hard parts

  • Offline by design. Every screen and rule had to work without a server.
  • Real conflicts. The same record can change in two places days apart.
  • The trade-off. Offline-first means every record can be edited in two places - sync with explicit conflict resolution costs UX complexity, but never silently loses a field worker's data.
07

My part

With the Auriga IT team, I built the offline-first Android app with sync and conflict resolution, and the central data and analytics layer behind the dashboards.

08

What I took away

  1. Design for your users' worst network.
  2. Put a human check between data and decisions.
  3. Speak the user's language, literally.

Want the detail behind the numbers - architecture, failure modes, what I’d change? Happy to walk through it.

Talk about this project →