evidoria

← Back to browse

Good practice Imported

Rocket Learning — AI-Graded WhatsApp Worksheets for Anganwadi Early Learners in India

India · New Delhi · See the India profile · See the New Delhi profile

Evidence: Descriptive / self-reported Top 16% 67/100 · Ask Evidence Copilot about this practice

Since 2020, Rocket Learning has used WhatsApp to deliver early-learning content to India's Anganwadi childcare network, training 1.3 million childcare workers and reaching 3 million children; an AI model built with Karya grades literacy/numeracy worksheets from photos with 97% ac

1300000
Anganwadi workers trained
97 %
Grading-model accuracy
1500000
Children covered by grading model
Rocket Learning — AI-Graded WhatsApp Worksheets for Anganwadi Early Learners in India

Details

Maturity
Scaling
Promoter
Rocket Learning (EKHO Foundation)
Period
2020–ongoing
Keywords
early childhood education, edtech, WhatsApp, AI/ML grading, workforce training

Context

Rocket Learning (EKHO Foundation), founded 2020 in New Delhi, sends personalised learning activities and coaching nudges to parents and Anganwadi workers over WhatsApp to strengthen India's public early-childhood-care system.

Results

The Star Educator Program has trained 1.3 million Anganwadi workers, reaching ~3 million children/parents across 10 states. A machine-learning grading model built with Karya, trained on 372,000 annotated worksheet images, reached 97% grading accuracy covering 1.5 million children.

Conclusions

Education Above All's own review notes 'further analysis will have to be carried out to understand the programme's entire impact.'

Implementation

Indicative cost
Low (< €50k)
Time to results
Medium (1–3 years)
Staffing & skills
Anganwadi workers (trained via Star Educator Program), ML engineering partner (Karya)

Conditions for success

  • high WhatsApp penetration among target households
  • large annotated-image training set for the grading model

Commonly funded by

Philanthropic / foundation funding

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful