Senior product engineer with 10+ years of experience shipping reliable web, mobile, cloud, data, applied-AI, and native macOS products from architecture through production verification.
I work best where correctness crosses boundaries: client UX, APIs, identity, persistence, cloud infrastructure, CI/CD, observability, and the real user flow all need to agree.
Local-first Flutter savings and goal tracker with manual accounts, recurring plans, versioned backup, optional owner-scoped sync, and review-before-write AI guidance.
- Flutter, Dart, Hive, Firebase Auth, Firestore, App Check, and Firebase AI
- no bank connection, money movement, or production credential in the repo
- cross-platform responsive UI with local ledger, backup, sync, and rules tests
Open 180-day practical robotics and ROS 2 field manual with deterministic visuals, interactive maths, portfolio projects, cloud progress, and a security-bounded AI tutor.
- Next.js 16, React 19, TypeScript, Firebase, Cloud Functions, and Gemini
- 180 static lesson pages with verified SVG and interactive maths models
- owner-scoped Firestore, App Check, quotas, replay protection, and server-side AI cost controls
- GitHub Actions, isolated emulators, and real Chrome desktop/mobile E2E
Live course · Source and setup
Privacy-first native macOS PDF audiobook reader with on-device OCR, neural narration, forced alignment, and synchronized word highlighting.
- Swift 6.2, SwiftUI, PDFKit, Vision, AVFoundation, MLX, and ONNX Runtime
- resumable, revision-pinned, SHA-256-verified model installation
- sandboxed read-only PDF access, local reading state, and bounded audio cache
- focused extraction, alignment, playback, and sentence-boundary regression coverage
- Web: TypeScript, Next.js, React, accessible responsive interfaces
- Mobile and native: Flutter, Android, Swift, SwiftUI, offline-first flows
- Backend and data: Node.js, Python, REST APIs, PostgreSQL, MongoDB, Firebase, Redis
- Cloud: Google Cloud, Cloud Functions, Firestore, Auth, App Check, CI/CD
- Applied AI: local inference, OCR, alignment, grounded model integration, provider isolation, deterministic validation, and cost controls
- Quality: real E2E, failure-state testing, security boundaries, observability, and evidence-backed releases
- Trace the complete flow before changing a production system.
- Keep credentials server-side and permissions fail-closed.
- Treat build success as a checkpoint, not proof that the user journey works.
- Prefer small, reviewable changes with executable contracts and real runtime evidence.
- Document setup, trust boundaries, limitations, and operational context.
I’m interested in senior product/software engineering work involving full-stack ownership, mobile or native platforms, cloud systems, developer infrastructure, applied AI, or robotics software. The repositories above are designed to be cloned, run, and evaluated—not just viewed as screenshots.



