Embedded ML Engineer – Edge Accelerator

Eurth Techtronics Pvt Ltd.; Hyderabad; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

Eurth Techtronics Pvt Ltd. is seeking an Embedded ML Engineer to design and deploy efficient ML models on real-world IoT systems with edge accelerators.

You will convert and optimize models for embedded platforms, integrate ML runtimes into RTOS/embedded Linux, and work with real-time sensors to achieve low latency and power-efficient inference. A strong background in C/C++ and embedded firmware is essential.

Qualifications

• BE/B.Tech/M.Tech in Electronics, Embedded Systems, CS or related discipline.
• 2–4 years in embedded ML, edge AI or firmware with ML inference.
• Strong C/C++ and basic Python; experience with RTOS/embedded Linux.
• Experience with TensorFlow Lite, ONNX Runtime or similar edge ML toolchains and quantization techniques.
• Git, Docker, cross-compilation toolchains (Yocto, CMake) and SWD/JTAG/GDB debugging.

Responsibilities

• Convert, quantize and compile TensorFlow, PyTorch or ONNX models to accelerator-specific formats.
• Apply post-training quantization, pruning, distillation and model slicing.
• Integrate ML runtimes (C/C++ or Python) into RTOS or embedded Linux firmware stacks.
• Handle real-time sensor inputs (camera, accelerometer, microphone) through inference engines.
• Profile and tune models for latency, memory, compute load and power.
• Build unit, integration and hardware-in-loop tests and reproducible benchmarking scripts.

Skills

C/C++

Python

RTOS/Embedded Linux

Education

BE/B.Tech/M.Tech in Electronics, Embedded Systems, CS

Tools

TensorFlow Lite

ONNX Runtime

Git

Docker

Yocto

CMake

Job description

We are seeking an Embedded ML Engineer to work on ML inference pipelines for low-power, real-time embedded platforms. You will design and deploy efficient ML models on custom edge AI accelerators in real-world IoT systems, covering model optimization, embedded firmware and toolchain management.

Key responsibilities

• Convert, quantize and compile TensorFlow, PyTorch or ONNX models to accelerator-specific formats
• Apply post-training quantization, pruning, distillation and model slicing
• Integrate ML runtimes (C/C++ or Python) into RTOS or embedded Linux firmware stacks
• Handle real-time sensor inputs (camera, accelerometer, microphone) through inference engines
• Profile and tune models for latency, memory, compute load and power
• Build unit, integration and hardware-in-loop tests and reproducible benchmarking scripts

Requirements

• BE/B.Tech/M.Tech in Electronics, Embedded Systems, Computer Science or a related discipline
• 2-4 years in embedded ML, edge AI or firmware development with ML inference integration
• Strong C/C++ and basic Python; experience with RTOS or embedded Linux
• Experience with TensorFlow Lite, ONNX Runtime or similar edge ML toolchains and quantization techniques
• Git, Docker, cross-compilation toolchains (Yocto, CMake) and SWD/JTAG/GDB debugging