Senior Embedded Edge AI Systems Architect

Location: New Delhi, India

Work Mode: On-site / Remote

Experience: 8+ years

Role Overview

We are looking for a Senior Embedded Edge AI Systems Architect to lead the deployment, hardware acceleration, and thermal and power optimization of large neural networks on resource-constrained embedded edge computing platforms.

The role focuses on integrating high-performance AI compute systems with real-time controllers, optimizing inference performance, and building reliable, deterministic sensor communication and synchronization pipelines for advanced robotics applications.

Key Responsibilities

• Embedded Compute Architecture: Architect dual-compute systems combining high-performance NVIDIA Jetson AGX Orin platforms (64GB) for vision and Vision-Language-Action (VLA) inference with real-time MCUs and x86 controllers for motor execution.
• AI Inference Optimization: Optimize neural network inference latency using TensorRT, TensorRT-LLM, FP8/INT4 quantization, CUDA, and kernel-level optimization techniques.
• Zero-Copy Communication: Design and optimize zero-copy shared-memory inter-process communication (IPC) pipelines between deep learning inference runtimes and ROS 2 middleware.
• Sensor Synchronization: Implement deterministic synchronization for 3D LiDAR, RGB-D and stereo cameras, and high-frequency IMUs using IEEE 1588 Precision Time Protocol (PTP) and hardware timestamping.
• Power and Thermal Management: Optimize onboard power distribution, dynamic thermal management, and NVIDIA Jetson MAX-N operating modes to maintain sustained performance under demanding workloads.
• System Reliability: Implement hardware watchdogs and robust embedded system monitoring to ensure reliable operation in real-time robotic environments.
• Hardware Integration: Drive the integration of embedded compute platforms, sensors, and real-time control systems into production-ready robotic hardware.

Required Qualifications

• Bachelor's or Master's degree in Computer Engineering, Electrical Engineering, Computer Science, or a related discipline.
• 8+ years of overall experience in embedded systems, systems architecture, or edge computing, with at least 5 years of hands-on experience in embedded systems development.
• Strong experience with NVIDIA Jetson platforms, CUDA programming, kernel optimization, and embedded Linux.
• Demonstrated experience deploying and optimizing quantized neural network inference pipelines on production robotic or embedded edge hardware.
• Proficiency in C++ and experience with Jetson Linux (L4T), TensorRT, TensorRT-LLM, and zero-copy shared-memory communication.
• Hands-on experience with IEEE 1588 PTP, hardware timestamping, deterministic sensor synchronization, and hardware watchdogs.
• Strong understanding of real-time systems, hardware-software integration, and performance optimization in resource-constrained environments.

Preferred Skills

• Experience with robotics platforms integrating multiple sensors and heterogeneous compute architectures.
• Familiarity with ROS 2 middleware and real-time communication between AI inference systems and motion-control components.
• Experience in thermal and power optimization for embedded AI systems.
• Strong system-level debugging and performance profiling capabilities across hardware and software layers.

---