Senior Mobile & Gen AI Engineer — I ship production AI that runs on the device, so user data never leaves the phone.
I bridge mobile and AI: React Native / Flutter apps where a real LLM runs entirely on-device via llama.cpp (llama.rn), with GPU/NPU offload and sub-2s cold start. 8+ years, 15+ consumer apps shipped for millions of users across fintech, transit, healthcare, and e-commerce.
- Migrated cloud inference → private on-device LLM inside a React Native + Flutter product: 4-bit ~1.8B model on a 4GB Android, p95 2.4s→380ms (6.3× faster), cloud serving spend −71%, accuracy tradeoff stated honestly.
- Measured on real devices (Firebase Performance traces), not a prototype. A senior mobile/AI screener at Property Finder confirmed this on-device-AI + senior-native-mobile combo is exactly what they screen for.
- Full teardown, architecture brief, and on-device latency metrics → https://github.com/Zulqurnain/on-device-llm-rn
- 📦 Proof repo (always-live hub): https://github.com/Zulqurnain/on-device-llm-rn
- 💼 LinkedIn: https://linkedin.com/in/zulqurnainjj
- ✉️ zulqurnainjj@gmail.com
- 🌐 Portfolio: https://zulqurnainj.com (intermittent — the proof repo above is the canonical, always-up surface)
Open to remote Senior AI + Mobile Engineer roles worldwide — on-device LLM, RAG, React Native / Flutter. If you're building AI features where data-localisation matters, let's talk.