Staff Engineer, Software

About the role

We are looking for a senior software engineer who can bridge two critical domains: AI/ML computer‑vision model development and native Android application engineering. You will own the end‑to‑end lifecycle of vision models for live camera‑feed monitoring and ship a production‑grade Android app that securely communicates with our cloud backend.

Key responsibilities

• Design, train, evaluate and tune computer‑vision models (detection, classification, segmentation, tracking) for multi‑camera live video, including dataset curation and rigorous benchmarking of accuracy, latency and throughput.
• Optimize models (architecture, quantization, pruning, distillation) to satisfy deployment constraints and re‑architect as those constraints evolve.
• Partner with operations/MLOps teams to maintain, upgrade and monitor models in production, handling regressions, data drift and continuous improvement.
• Develop and ship a native Android app (Kotlin, Jetpack Compose, MVVM/Clean Architecture, Hilt, Coroutines/Flow, Room, WorkManager) that integrates securely with cloud services via REST/gRPC, OAuth2/JWT, TLS and FCM.
• Implement Android CI pipelines, unit and instrumentation testing, crash reporting and Play Store release processes.

Required profile

• 5+ years of professional software engineering experience.
• Proven track record delivering vision‑based models to production and shipping Android apps to the Play Store.
• Experience with cloud AI training platforms such as AWS SageMaker, GCP Vertex AI or Azure ML.
• Strong collaboration skills with deployment and operations teams and a product‑mindset focused on measurement and optimization.

Required skills

• Python, PyTorch, TensorFlow, TensorFlow Lite, ONNX, YOLO, DETR, ByteTrack.
• Kotlin, Android Jetpack, Compose, MVVM, Clean Architecture, Hilt, Coroutines, Flow, Room, WorkManager.
• REST, gRPC, OAuth2, JWT, TLS, FCM, Android Wi‑Fi and BLE APIs.
• Experience with model optimization techniques (quantization, pruning, distillation) and on‑device inference acceleration (NNAPI, Qualcomm SNPE).