BTS - Back to School Planner Assistant

#devchallenge #herokuchallenge #webdev #ai

This is a submission for the Heroku "Back to School" AI Challenge

What I Built

BTS - (Back to School) is an intelligent, multi-persona planning assistant that transforms the chaotic back-to-school experience into a structured, personalised journey. The application addresses one of the most universally stressful periods in education by providing AI-generated, phase-based transition plans tailored to three distinct personas: Students, Parents, and Educators

The Problem It Solves

Every year, millions of families and educators face the same challenge: how to successfully transition from summer break back to the structured demands of the academic year. This isn't just about buying supplies—it's about:

Students: Adjusting sleep schedules, managing anxiety, building study routines, and preparing socially and academically
Parents: Coordinating family logistics, supporting children emotionally, establishing communication with schools, and managing multiple children's different needs
Educators: Setting up classrooms, preparing curriculum, building relationships with new students, and maintaining work-life balance

Traditional approaches rely on generic checklists that ignore the unique challenges each persona faces at different stages of the transition process

Solution

BTS creates intelligent, contextual plans that adapt based on:

Persona-specific expertise: Each role gets plans designed around their unique challenges and priorities
Phase-based timing: Activities change based on whether you're 2 weeks out, 1 week out, in the first week, or building habits in the first month
Personal customisation: Users can specify focus areas (Time Management, Organisation, Social Skills, etc.) and custom requirements
Interactive management: Plans aren't static—users can edit, delete, and customise activities in real-time

Demo

Frontend: https://bts-ui-9418e8cc32cc.herokuapp.com/
Backend API: https://bts-onboard-21c136a3dfd0.herokuapp.com/
Github Repo: https://github.com/Nicanor008/Heroku-Back-to-School-AI-Dev-Challenge

How I Used Heroku AI

Heroku Managed Inference Integration

Primary AI Engine: Claude-4-Sonnet via Heroku Inference API
Endpoint: https://us.inference.heroku.com/v1/chat/completions
Model: claude-sonnet-4 for optimal reasoning and planning capabilities

The application implements a sophisticated multi-agent system where different AI agents specialise in different personas to achieve the planning

Agent Coordination Strategy

Dynamic Context Building: Each agent receives rich context, including:

Phase-specific challenges (preparation vs. adjustment vs. habit formation)
User customisations (focus areas, special requirements)
Persona expertise (age-appropriate, role-specific terminology)
Temporal awareness (timeline-sensitive activity suggestions)

Technical Implementation

Agent Specialisation Pattern: Instead of a single generalist AI, BTS implements specialised agents with deep domain knowledge
Context-Aware Prompt Generation: Each agent builds sophisticated prompts