amansultan
scaling lesson production through a modular editor
Amansultan is a licensed platform for schools where teachers present interactive lessons in class. By 2025, the product contained ~700 lessons, but production didn’t scale and teacher adoption remained low.
The bottleneck wasn’t visual design — it was the monolithic lesson editor and its workflow.
edtech · internal tools
scalability broke
The production workflow functioned at small volumes but collapsed under scale.
Each lesson required coordination across multiple roles, while the editor itself introduced friction instead of reducing it.
Changes in one part caused rework across the pipeline. A single lesson took days to assemble.
modular editor architecture
Both legacy editors were monolithic. All mechanics lived inside a single interface with shared logic and settings. Adding or modifying one mechanic required changes in the core, increasing complexity and slowing development.
To solve this, I proposed restructuring the editor into a modular, template-based system. Each slide begins by selecting a mechanic template, opening a focused editing mode instead of a single overloaded interface.
editing in practice
Each slide begins with selecting a mechanic template.
Instead of configuring everything in one overloaded panel, the editor opens a focused editing mode specific to that mechanic.
mechanics pack
To unblock content production, scope was intentionally limited. We defined an initial pack of eight mechanics (including test, matching, and sorting) to enable faster assembly, predictable quality, and a repeatable production pattern.
trade-offs
This decision reduced flexibility: fewer unique slides, limited mechanic variety, and dissatisfaction from stakeholders expecting faster breadth.
I explicitly prioritized speed, consistency, and scalability over flexibility. Given that each new mechanic previously required weeks of development, this trade-off was necessary to establish a scalable foundation.
results
The editor shipped to production and became the primary way lessons are assembled. The team’s focus shifted from
“we need more mechanics”
to
“we need better content.”
This became especially visible in primary school mathematics, where several lessons could be produced per week.
reflection
Looking back, I would invest more time in redesigning the editor UI. I also see the next step in moving from strict templates toward controlled free editing, which I intentionally postponed to first establish a scalable foundation.
Without restructuring the editor, Amansultan wouldn’t have been able to scale lesson production.