Description :

We are looking for an experienced Edge AI Architect to design, optimize, and deploy AI/ML models on edge computing platforms including CPUs, MCUs, DSPs, GPUs, and NPUs. This role requires expertise in embedded AI, TinyML, edge inference acceleration, model optimization, and hardware-aware.

Responsibilities :

- Architect end-to-end Edge AI pipelines for heterogeneous hardware (CPU, MCU, DSP, GPU, NPU).

- Optimize models with quantization, pruning, distillation, compression, and mixed precision.

- Deploy AI models using TensorRT, OpenVINO, ONNX Runtime, TFLite, TVM, SNPE, Vitis AI.

- Benchmark and optimize on ARM Cortex, RISC-V, NVIDIA Jetson, Intel Movidius, Qualcomm Snapdragon, NXP i.MX, TI Jacinto, EdgeTPU.

- Integrate AI workloads with Linux, RTOS, Android, and bare-metal systems.

- Ensure secure model deployment (encryption, licensing, tamper-resistance).

- Collaborate with hardware, firmware, and software teams to deliver real-time inference.

- Research edge accelerators, compiler toolchains, federated learning, TinyML, AIoT.

Qualifications :

- Bachelor's/Masters/Ph.D. in Computer Science, Electrical Engineering, Embedded Systems, or AI/ML.

- 10+ years of experience in embedded AI, deep learning, and edge computing.

- Strong in C, C++, Python, RTOS, and embedded firmware.

- Proficient in TFLite, ONNX, OpenVino etc.

- Expertise in hardware-aware training, compiler optimizations, and inference acceleration.

Preferred Skills :

- Experience with biometrics, computer vision, speech, NLP on embedded devices.

- Familiarity with TinyML, federated learning, AIoT, neuromorphic computing.

- Knowledge of MLOps, LLMOps, Docker, Kubernetes for edge AI orchestration.

- Exposure to FPGA-based AI acceleration and hardware-software co-design.