FocusKPI is seeking an Android AI/ML Engineer - On-Device to join one of our clients, a high-tech SaaS company.
The client is looking for a highly capable Android AI/ML Engineer - On-Device to help build intelligent, privacy-first mobile systems that can detect, respond to, and learn from dynamic real-world conditions. This role involves deploying resource-efficient ML models directly on Android devices, combined with backend integration for model management, telemetry, and secure update delivery.
The ideal candidate has a strong background in on-device intelligence and cloud-integrated systems, especially in applications that require responsiveness, adaptability, and strict privacy controls.
Work Location: Mountain View, CA (on-site 4 days a week)
Duration: 6-month contract; hybrid role
Pay Range: 80/hr to $93/hr

Responsibilities:

• Design, develop, and deploy on-device machine learning models optimized for Android, ensuring low latency and minimal resource consumption.
• Build robust and scalable ML pipelines using Android-native frameworks such as:
• TensorFlow Lite
• ML Kit (including GenAI APIs)
• MediaPipe
• PyTorch Mobile
• Build robust and efficient on-device data pipelines and inference mechanisms for real-time decision-making.
• Apply model optimization techniques such as quantization, pruning, and distillation for performance on mobile hardware.
• Ensure privacy-first design by performing all data processing and inference strictly on-device.
• Collaborate with backend teams to integrate with cloud-based model orchestration systems (e.g., MCP or similar) for:
• Model versioning, delivery, and remote updates
• Telemetry collection and model performance monitoring
• Rollout and A/B testing infrastructure
• Implement secure local storage, encrypted data handling, and telemetry pipelines that meet privacy and compliance standards.
• Support adaptive model behavior through on-device fine-tuning, personalization, or federated learning workflows.

Qualifications:

• 5-7 years of experience with a Master’s degree, 3+ years of experience with a PhD
• A candidate with a strong on-device ML experience is highly prioritized and required
• A deep understanding of app architecture, background processing, and system APIs.
• Hands-on experience with on-device ML frameworks: TensorFlow Lite, ML Kit, MediaPipe, PyTorch Mobile.
• Solid understanding of mobile performance optimization, including model size, memory usage, and latency.
• Experience with signal processing and real-time ML logic
• Proven ability to integrate Android apps with backend/cloud systems for: Model lifecycle management (delivery, updates, rollback), Logging, telemetry, and analytics
• Experience with secure Android development, including permissions, sandboxing, encryption, and local data protection.
• Strong understanding of privacy-first ML system design and local-only data processing.
• Experience in Android development using Kotlin and/or Java is nice to have
• Documentation is essential

DESIRED QUALIFICATIONS:

• Specialized experience in catering to mobile devices. Mobile is a very resource constrained system.
• Prior experience in optimizing ML solutions on on-device.
• Experience in quantifying or optimizing on-device and handling privacy like federated learning or something similar is preferred.
• Experience in TensorFlow Lite (TFLite), Edge device, and mobile device optimization.
• Experience working with model orchestration platforms (e.g., MCP, Vertex AI, SageMaker, or internal tools).
• Familiarity with federated learning, on-device personalization, or differential privacy.
• Experience with anomaly detection and behavioral modeling
• Experience with secure telemetry and backend integration
• Background in building real-time, data-driven features in mobile apps at scale.
• Model optimization for ARM/mobile hardware
• Familiarity with cloud infrastructure (e.g., GCP, AWS) for ML model deployment and monitoring.
• Previous work in high-sensitivity domains such as identity, privacy, mobile security, or regulated industries is a plus.
  No C2C resumes are considered, and avoid Android developers for this role.
  Thank you!
  FocusKPI Hiring Team
  Founded in 2010, FocusKPI, Inc. (FocusKPI) is a data science and technology firm specializing in predictive analytics practice and methodologies. FocusKPI is a US company headquartered in Silicon Valley, California, with an East Coast office in Boston, Massachusetts.
  NOTICE: Please be aware of fraudulent emails regarding job postings, job offers and fake checks. FocusKPI’s recruiting team will strictly reach out via @focuskpi.com email domain. If you have received fraudulent emails now or in the past, please report it to https://reportfraud.ftc.gov/ .
  The domain @focuskpijobs.com is fraudulent and not related to FocusKPI. Please do not not reply or communicate to anyone with @focuskpijobs.com.
  5Jr3PkXx1

• Design, develop, and deploy on-device machine learning models optimized for Android, ensuring low latency and minimal resource consumption.
• Build robust and scalable ML pipelines using Android-native frameworks such as:
• TensorFlow Lite
• ML Kit (including GenAI APIs)
• MediaPipe
• PyTorch Mobile
• Build robust and efficient on-device data pipelines and inference mechanisms for real-time decision-making.
• Apply model optimization techniques such as quantization, pruning, and distillation for performance on mobile hardware.
• Ensure privacy-first design by performing all data processing and inference strictly on-device.
• Collaborate with backend teams to integrate with cloud-based model orchestration systems (e.g., MCP or similar) for:
• Model versioning, delivery, and remote updates
• Telemetry collection and model performance monitoring
• Rollout and A/B testing infrastructure
• Implement secure local storage, encrypted data handling, and telemetry pipelines that meet privacy and compliance standards.
• Support adaptive model behavior through on-device fine-tuning, personalization, or federated learning workflows