Projects2026

Academic Ally

Overview

An agentic study platform: students upload course syllabi and it auto-plans their semester, with built-in focus sessions and integrated task and project management. Powered our Round 1 submission into the top 10 of the KPMG Academic Innovation Challenge.

Role: Full-stack developer
Context: KPMG Academic Innovation Challenge
Input: Course syllabus
Result: Top 10 submission

Results at a glance

Top 10 KPMG placement
Syllabus → Plan Input
End-to-end Scope

Problem

Students start every semester re-deriving the same plan by hand: reading each syllabus, mapping deadlines to a calendar, and stitching together separate apps for tasks, focus, and project tracking. The setup cost is high and the plan goes stale the moment a course changes.

Approach

Academic Ally turns a syllabus into a semester. Students upload their course outlines and an agentic system extracts milestones, deadlines, and workload, then generates a plan with built-in focus sessions and integrated task and project management — so planning, doing, and tracking live in one place instead of three.

Technology

Next.js · React · Cloudflare D1 · Cloudflare R2

Technical decisions

Syllabus parsing as the core primitive

We treated the uploaded syllabus as the single source of truth — files land in Cloudflare R2, structured plan data in D1 — so every downstream feature (focus sessions, tasks, projects) derives from one parsed representation rather than asking the student to re-enter their schedule.

Cloudflare-native data + storage

Built on Cloudflare D1 and R2 to keep the whole stack on one edge platform, which kept latency low and let a small team ship a working end-to-end submission inside the challenge timeline.

Tradeoffs

Auto-planning vs. student control

Auto-generating the plan removes setup friction but risks over-prescribing. We leaned toward an editable generated plan so students keep the final say rather than fighting a rigid schedule.

Breadth of integrated features under a deadline

Bundling planning, focus, and task management into one Round 1 submission meant each surface was intentionally lightweight. It proved the concept top-10, but depth on any single surface was deferred.

Outcomes and lessons

Built an end-to-end syllabus-to-plan workflow that reached the competition top 10.
Kept generated plans editable so students could correct or reshape the proposed schedule.
The next release would deepen planning controls and validate schedule quality with students.

Related projects

Telecom Churn Prediction

An end-to-end churn analysis for a telecom operator, read as one narrative: EDA to discover the business problem, retention segmentation, an XGBoost risk model, and a 'Renewal Rescue System' proposal — framed through a Philippine/ASEAN market lens.

Read case study View project

Meera

An agentic AI front door for university service desks: concerns arrive in plain language, get parsed and classified, are resolved automatically where possible, and routed to the right team when not. Top 3 of 100+ teams at the KPMG Academic Innovation Challenge.

Read case study

Mayumi Cards

A Japanese flashcard app built around a client-side SRS algorithm. The hard part wasn't the UI — it was tuning the review-interval policy so daily study actually compounds.

Read case study View project

Benkyō

A spaced-repetition Japanese learning platform built for Ateneo's Introduction to Japanese 11 class. Schedules vocabulary reviews around retention patterns and tracks daily study streaks — serving 50 real students.

Read case study View project