Jonathan Odul

Leadership and organizational impact

Management & Alignment: Leading a bilingual team (6 engineers + offshore contractors) responsible for a platform securing 100,000+ devices (iOS, Android, Windows). Central point of contact (Product, QA, Support, Management).

Crisis Management: Resolving a critical incident (530 devices blocked) with RCA and a fix deployed in <3h (on a Friday evening). Created the post-mortem model that became the company standard.

Delivery & HR: Managing monthly releases (backend/admin) with strict migration controls. Recruitment, coaching (1:1), evaluations, and team performance management.

Partnerships: Achieving Google Android Enterprise Silver partner status. Sole representative at Android Partner Day APAC 2025 (Bangkok); generating business opportunities with Google Tokyo.

AI Strategy: Leading AI adoption (~30 people): budget, training (EN/JP), usage charters, and creating an AI-driven Android regression testing framework (with production safeguards).

Technical scope and delivery (hands-on)

Critical Development: Resolving race conditions during enrollment (Django), Wi-Fi/anti-reset policies (Android), hardening Kiosk mode (Android 13-16).

Scalability: Secure deployment of 100k+ devices (up to 5,000/h), optimizing push gateway to 1,000/min (AWS Kinesis, DynamoDB deduplication).

DevOps & CI/CD: AWS CodePipeline (Elastic Beanstalk deployments), IAM security (least privilege), one-click Django migrations via AWS SSM.

Quality & Modernization: Upgrading Python 3.12 and Django 5.2 LTS. Improving test coverage from 89% to 91% (+500 tests).

Large Account Integrations: Microsoft Intune, Apple VPP/ABM and DDM, mass enrollment (QR, Zero-touch), large-scale policy management (up to 40,000 devices).

I led the development of Xloudia, an image/video recognition platform using keypoints with augmented reality (classic computer vision, pre-deep learning): matching live camera feeds with a database of pre-recorded images, distributing it (sharding) across a horizontally scaled worker pool and returning the first match, so that adding nodes reduced recognition latency; serving thousands of requests per day on AWS.

I built interactive computer vision installations (facial recognition, body and limb tracking) with Unity, C#, C++ and OpenCV.

Co-authored the peer-reviewed publication “Neuron Analyzer Suite”, improving automated neuron detection, which is now used by Japanese life science laboratories.