About the role
About
We are reimagining the phone. AI phones need to be proactive, intuitive, intelligent and low cost. This is a mission critical for rest of the world to adopt intelligence.
There are over 5 billion people using everyday applications such as email, messaging, notes, tasks, and calendars, yet most are not AI-native. Our mission is to build proactive applications and devices that bring intelligence into daily conversations, coordination and workflows, with minimal prompting.
Responsibilities
* Evaluate suitable open-weight and commercially available models for on-device execution.
* Benchmark model performance on target mobile hardware.
* Integrate models into Android applications and system services.
* Optimize inference latency, memory consumption and power usage.
* Evaluate quantization, compression and model distillation techniques.
* Work with CPU, GPU and NPU acceleration.
* Evaluate local model capabilities for different workflows.
* Build hybrid local/cloud model routing mechanisms.
* Identify which tasks can be handled locally and which require cloud inference.
* Work with AI agent engineers to integrate local models into the agent runtime.
Requirements
* Strong ML systems or inference engineering experience.
* Experience deploying and optimizing models.
* Strong Python and C++ skills.
* Understanding of neural network inference.
* Experience with model quantization and performance optimization.
* Understanding of hardware constraints and inference benchmarks.
Nice to have
* Experience with on-device LLMs.
* Qualcomm QNN, MediaTek NeuroPilot or similar acceleration frameworks.
* ONNX Runtime, LiteRT, TensorFlow Lite or ExecuTorch.
* Android NDK.
* Mobile or edge AI deployment.
* Multimodal model optimization.
Originally posted on Himalayas