Pillar Four · For institutions

Million-dollar problems. Not million-dollar budgets.

Most of the world’s students are taught by institutions the enterprise-software industry politely ignores: regional publics, community colleges, polytechnics, and the fast-growing campuses of India and the Gulf. The platforms they rent are priced for someone else and tuned for no one in particular. We think the economics of that just changed.

Become a design partner
The thesis

Software finally got cheap enough to fit you.

A learning management system runs $5 to $30 per student per year before implementation, and the implementation services often cost more than the first-year license. What that buys is software built once, for everyone, which means tuned for the average of thousands of campuses. Your bilingual student body, your evening-heavy schedule, your accreditation quirks: edge cases, priced accordingly.

AI-era development changes the math. Tools that took a vendor a product team and a year can now be built, customized per campus, and maintained by a very small team. To be clear about what we are and aren’t proposing: nobody needs a rip-and-replace pitch. These are nimble layers that ride alongside the LMS you already own, integrate through the standards you already use, and get shaped to your campus the way only bespoke software used to be.

The toolkit

Three tools, each with published evidence behind it.

Gateway courses

Adaptive courseware

The intro courses that decide retention. Adaptive redesigns have lifted success rates by around 10 points in published institutional work. Ours plugs into the same engine as our student companion, so the tool that helps your student also informs your instructor.

Achieving the Dream ↗
Retention

Early alert & recovery

Engagement signals flag struggling students before the midterm, with published accuracy near 85 percent. The same literature is blunt that pure automation underperforms, so this augments your advisors and never replaces them.

EDUCAUSE ↗
Academic integrity

Assessment redesign tooling

AI made take-home assessment a genre problem. We implement the emerging consensus (process-based work, staged checkpoints, explicit per-assignment AI rules) as software faculty adopt course by course, instead of a policy PDF nobody operationalizes.

AI Assessment Scale ↗
Where and when

Sequencing, honestly.

This pillar comes after our consumer tools prove the engine. That is deliberate, and it is why this page makes no sales pitch. The founder teaches at exactly the kind of institutions this pillar serves, and the first pilots will run where the problems are personally known.

Geography is part of the design, not an afterthought. India’s tier-2 and tier-3 colleges and polytechnics are digitizing fast under NEP 2020 with budgets no US vendor takes seriously, and the Gulf’s private campuses want customization the incumbents can’t economically offer. Tools built to be reshaped per campus are the first ones that can honestly serve both.

Design partners

If the problems on this page are your problems, we’d like to design with you, not at you. Write to richa@thinkathena.ai with “design partner” in the subject line.

Stay close

Follow the work as it becomes real.

Essays on assessment, retention, and AI-era teaching land in the newsletter first, alongside build notes from the consumer pillars proving the engine this one will inherit.

Keep me posted

One email does it: no forms, no spam, no tricks.