Notes from the back of a classroom

This article is authored by Jayasree Ramesh, COO, education, Kotak Education.

On a recent visit to a government school in rural Maharashtra, I watched a Class 2 teacher run something her children simply call "the train." Her classroom floor was divided into six stations, each no bigger than a mat: A Focus Station where children settled in with a quiet listening game, a Story Station with a picture book and a child reading aloud to two others, a Skill Drill Station for practising letter sounds, and three more for numbers, writing, and open play. Every few minutes she clapped twice, called out "chuk-chuk," and a small group of children stood, joined hands, and chugged, quite literally, to the next station, while she moved down the row like the engine pulling her own train, stopping at each bogey just long enough to see who needed her. No child sat still for long. No two children were being taught in the same way at exactly the same moment.

NEP (Representative photo) (HT_PRINT)

I have thought about that classroom often since. NEP 2020 turned six this year, and when it began, this is what it asked for: that every child learn to read and count with confidence before school demanded anything more of them. NEP 2020 gave Indian school education a clear, coherent architecture, a curriculum, a mission for foundational learning, a teacher-training platform, an assessment philosophy--articulating pathways to skilful execution on-ground, shaping positive growth mindsets, giving direction. More than 2.5 million teachers have been trained under its NISHTHA-FLN module, with instruction purposefully shifting from rote drilling to play-based, activity-led pedagogy. The results are showing. The share of Class 3 children in government schools able to read a Class 2-level text rose from 16.3% in 2022 to 23.4% in 2024, the best such movement in 20 years of tracking.

Enhanced technological pedagogical competence and teachers developing fluency in AI have also contributed to the upswing in learning outcomes. Teachers across Maharashtra now have access to an AI-enabled companion, built in Marathi with the state's own teacher training body and delivered through DIKSHA. It helps them plan exactly this kind of differentiated station-wise teaching without spending their evenings doing it from scratch. A teacher using it told me it gives her back the one thing she never has enough of: Time to actually look at each child, instead of planning six versions of the same lesson at midnight. Multiply her by the tens of thousands of teachers now using similar tools, and you start to see what Artificial Intelligence (AI) in our schools should mean--not a machine replacing a teacher, but one quietly doing the preparation so she can do the part only she can do.

The honest part of NEP’s sixth anniversary is acknowledging the gaps alongside the wins. Reading proficiency improves in the early years but drops as students move to higher grades. Barely 6 out of 10 children who start Class 1 make it to Class 10. Fewer still, finish Class 12. Secondary school is the weakest link in the chain. Secondary GER, flat since 2020-21, stood at 78.7% in 2024-25; Higher Secondary trails further behind at 58.4%--both short of NEP's 100%-by-2030 target. Behind this sits a teacher-supply constraint: secondary and higher-secondary vacancies nearly doubled in a single year, from 2.3 lakh to over 4 lakh, at precisely the stage where subject-specialist teachers matter most. Nationwide, more than a lakh schools still run on a single teacher. NEP's goal of acquainting half of all school students with vocational skills by 2025 has remained unmet, limiting school-industry linkages. Children from underserved communities drop out of school due to economic hardship and the need to do paid labour to support their families, gender disparities, long travel distances to secondary school, inadequate basic infrastructure or teaching staff. Education spending, still short of the 6% of GDP the policy hoped for, hasn't grown as fast as our ambition.

This gap is what partnerships between government, philanthropy, and CSR-backed foundations have quietly begun to close--not by replacing public investment, but by reaching the corners a stretched budget cannot. It is, I think, the most underappreciated part of this story: a teacher-training platform here, a diagnostic tool there, a foundation absorbing the cost of piloting something which a state government scales up later, to achieve rewarding outcomes.

So, what should a reader actually take from six years of NEP 2020? If you're a parent, ask your child's school a different question this year — not whether the syllabus was completed, but whether your child can read and solve a problem on their own. If you're a teacher, know that the 50 hours of training NEP - promises you annually is a right, not a favour, and it's worth asking for. And if you simply care about where it is headed, don't judge the policy by the country's average--ask what your own state's foundational learning and retention numbers say. They are public and shine a light on learning outcomes actually attained.

I still think of that classroom in Maharashtra: six small stations, a teacher playing engine to a train of seven-year-olds who have no idea they are being taught differently than any generation before them. NEP 2020 is not finished yet and won't be by any single anniversary. But for the first time in a long while, Indian classrooms are being built to meet each child roughly where she is, rather than where the syllabus assumes she should be. That is worth six years of work. It is worth the next six too.

(The views expressed are personal)

This article is authored by Jayasree Ramesh, COO, education, Kotak Education.