Building and launching an AI-first classroom app
Anyone can download the app and create a teacher profile.
Situation
Teachers in Indian schools prepare lessons, set and grade assignments and compile reports largely by hand, and school records sit across registers, separate platforms and teachers' memory. Most AI tools give teachers one more thing to manage.
Approach
Led product and operations while our CTO built the platform. EduWand connects the whole job: what a teacher prepares feeds the assignment, and both feed the report. I set up the company, ran go-to-market and tested pricing in conversations with schools and distributors.
Outcome
EduWand is live on iOS and Android. We ran four AI-training workshops for 50+ teachers and reached commercial negotiations with an 850-student school. We wound down EduWand in Oct 2026 after cash ran out before school interest converted to revenue.
Launch video
Plays from YouTube when you press play, or watch it on YouTube ↗
What EduWand does
An AI-native platform for teachers on web, iOS and Android. The teacher chooses the material and stays in control of everything that goes to students.
Lesson Studio
The teacher gathers sources for each class and topic, and EduWand turns them into lesson plans, slides and study material.
Assignment Lab
Builds an assignment from what was taught in the topic. Each question is tagged to a topic and a Bloom's taxonomy level, and the teacher reviews it before it goes out.
AI Wand
A built-in assistant opened by tapping the cat mascot. It answers questions about the teacher's own classes, assignments and grading, and prepares changes for the teacher to confirm.
Attainment Report
Shows whether the class learned what the curriculum says it should, by topic and by student.
The app
Timeline
- Jun 2025Started EduWand with a colleague, alongside our roles at TresVista. I'm a third-generation educator: I grew up watching my mother and grandmother bring school admin home, and saw my mother struggle to use AI tools to make teaching material.
- 2025The first partnership ended over a mismatch in skills, and a second co-founder did not work out. Aneesh then joined as CTO and built a web app in three weeks.
- 2025Schools agreed to try the web app, but we couldn't onboard any. We learned that schools hold sensitive student data, don't trust unfamiliar software and want one complete system rather than several apps.
- Apr 2026Incorporated LessonForge LLP, gained DPIIT recognition and applied for government grants.
- Apr 2026Started building the mobile app and our go-to-market, wrote our brand guide, ran our social channels and secured $5K in AWS startup credits.
- Jul 2026Went full-time on EduWand after a 3-month notice period at TresVista.
- Aug 2026Won best pitch at a Bangalore pitch competition. When a scheduled speaker dropped out, St. Francis de Sales, Nagpur, asked me to run their teacher training; I built the workshop in under two hours. The school paid ₹1,000 for the session, our first income, and featured it in its magazine.
- Before launchFiled trademarks. Apple approved the iOS app in three days and Google approved the Android app in six.
- 6 Oct 2026Launched at the Education Supply and Franchise Expo (ESFE 2026) and spoke with 250+ educators, software distributors and school owners. The day-one price was rejected, so we tested new price points over the next two days and found a workable starting price. Refocused the product on school administration, since schools make the buying decision, and planned the next version as an AI co-ordinator that would produce reports in each school’s standard format at any point from live classroom data.
- 2026Ran paid ads for teacher workshops (50+ leads in a week at about ₹40 each) and held two online workshops, which 18 teachers joined from about 50 sign-ups.
- Oct 2026Ran an AI workshop for teachers at an 850-student school in Bangalore, which then entered commercial negotiations with us.
- Oct 2026Wound down after cash ran out before school interest converted to revenue.
What I learned
- Build a solution that will sell, not only one that solves a single problem.
- Focus on one pain point and break it into parts.
- B2B sales in EdTech are longer and more expensive than expected, and schools buy through distributors and consultants they trust.
- Many school administrators prioritise admissions over teacher workload, so the pitch has to start with the administration.
- AI can change how a school runs, if the school is open to change.
What's next
I don't plan to restart EduWand, even if funding becomes available. I want to work alongside an experienced founder who has built companies past $10M ARR, in a founder's office role where I can add value across the business.
On the ground
