EdTech and e-learning development
Learning platforms built by people who also run a school of their own.
We have an unusual position here: we run Codexa Academy, and the platform it runs on is one we built. Enrolment, live classes, assignments, auto-graded quizzes, a gradebook and certification are not a capability list for us — they are software we operate, support and get complained about when it breaks. That tends to change what you build.
The problems specific to edtech
Completion, not enrolment
Signing learners up is easy and worth little. Everything that matters happens in the gap between starting and finishing, and that gap is a product design problem more than a content one.
Assessment at scale
Auto-marking is straightforward for multiple choice and hard for anything worth assessing. Most platforms need both, plus a sane path for human marking.
Three different users
Learners, instructors and administrators want incompatible things from the same system. Designing for one at the expense of the others is the most common failure here.
Live and recorded together
Cohort teaching means scheduling, attendance, recordings and materials all have to hang off the same structure.
- Learning platforms with courses, modules and lessons
- Cohort and live-class management with attendance
- Assignment submission and grading workflows
- Auto-graded quizzes and gradebooks
- Certification and credential issuing
- Enrolment, payments and student onboarding
- Children's data
- Teaching under-18s brings a stricter regime — consent, minimisation, retention and safeguarding — and it constrains the data model rather than sitting beside it.
- Accessibility
- Education has both the strongest legal expectations and the widest range of learner needs. Captions, keyboard access and screen-reader support are baseline.
- Video delivery
- Hosting, transcoding and bandwidth are a recurring cost that scales with success, and it belongs in the business model.
- Academic integrity
- Proctoring, question banks and submission checks are their own scope, and worth deciding on before launch rather than after an incident.
What we have shipped in edtech
Should we build a platform or use an existing LMS?
Use an existing one unless your teaching model does not fit it. Established platforms handle course delivery, quizzing and reporting competently for a subscription. Building is justified when your model is unusual — mixed live and asynchronous cohorts, bespoke assessment, or a credential that has to integrate elsewhere — or when the platform is the product you sell.
How do you handle live classes?
Generally by leaning on an existing video service and building the structure around it: scheduling, attendance, materials, recordings and the follow-up. Building video delivery itself is rarely worth it; the value sits in everything attached to the session.
Can certificates be verified?
They should be. A certificate nobody can check is decoration, and a public verification page turns each one into something a graduate can link from a professional profile. It is a small feature with disproportionate value to both the learner and the institution.
What actually improves completion rates?
Cohorts with real deadlines, visible progress, prompt feedback and a person who notices when someone goes quiet. The strongest levers are structural rather than technical — which is worth knowing before spending a budget on features that will not move the number.
Get an indicative range for a project of this shape before you talk to anyone.
Estimate the costLet's build something that lasts.
Tell us what you're working on. We'll come back within two business days with a clear next step — not a sales script.