Upload and Query
- 1 Open app
- 2 Upload PDF or text document
- 3 Document processed and embedded on-device
- 4 Ask a question in natural language
- 5 Get answer with source references — all offline
EduMate Lite
In ProgressOn-Device AI Learning Companion
A privacy-first mobile learning companion that runs entirely on-device. Upload documents, ask questions, and get AI-powered answers — all without internet. Demonstrates edge AI architecture with on-device RAG pipeline using Gemma small language model and ObjectBox vector storage.
// screenshots
// features
What it does
Fully On-Device RAG
Complete RAG pipeline running locally — no cloud, no API calls, no data leaves the device.
Gemma Small Language Model
Google's Gemma SLM optimized for mobile inference with low memory footprint.
ObjectBox Vector Storage
High-performance embedded vector database for on-device similarity search.
Zero Network Dependency
Works in airplane mode, offline environments, and restricted networks.
Document Upload & Processing
Upload PDFs and text files, automatically chunked and embedded on-device.
// architecture
Under the hood
On-device pipeline with embedding engine, vector store, and SLM inference — all running within Flutter.
Document Processor
Dart
File parsing, chunking, and preprocessing
Embedding Engine
TensorFlow Lite
On-device text embedding generation
Vector Store
ObjectBox
Embedded vector database for similarity search
Inference Engine
Gemma via TFLite
On-device language model for answer generation
// user journeys
How it gets used
// tech stack
FlutterGemmaObjectBoxDart
🧠
Phoenix
Ready when you are
Hey! I'm Phoenix — I know Titas's work, projects, and experience. Ask me anything — from distributed systems to production RAG, or what it's like building at Tesco and VMware.