Time-Series ML Engineer for Edge Inertial Sensors

siwaresystems; Riyadh; On-site; SAR 180,000 - 280,000; Full time

Job summary

siwaresystems in Riyadh is building a Systems-Level Integration team focused on smart, high-performance inertial sensors. As a Machine Learning Engineer in our Riyadh office, you will pioneer deep learning to enhance MEMS IMU sensor performance.

In the initial phase, you will focus on software, simulation, and data-driven modeling using datasets from our core engineering group. You will train models to denoise signals, fuse data, and correct stochastic errors, with cross-border collaboration to

Qualifications

• BSc or MSc in a related engineering field.
• Saudi Council of Engineers Engineer Grade membership is mandatory.
• Strong hands-on experience with deep learning and time-series data.
• Proficient in Python with good software engineering practices (unit tests, modular design).
• Git proficiency is mandatory.

Responsibilities

• Design, train, and validate neural networks for denoising and fusion of inertial data.
• Develop AI models for virtual sensing of MEMS IMU outputs.
• Build robust data pipelines for high-frequency inertial datasets.
• Collaborate with Egypt-based teams to ensure models fit edge devices (TinyML).
• Prototype ML approaches from research papers and optimize models for embedded targets (ARM Cortex-M).
• Regularly survey state-of-the-art ML research at the intersection of ML and inertial systems.
• Quantize and prune models for deployment on resource-constrained hardware.

Skills

Python proficiency

Time-series data

Deep learning

Git proficiency

Unit testing

Education

BSc or MSc in Computer/Electrical/Aerospace Eng

Tools

PyTorch

TensorFlow

ONNX

Job description

siwaresystems in Riyadh is building a Systems-Level Integration team focused on smart, high-performance inertial sensors. As a Machine Learning Engineer in our Riyadh office, you will pioneer deep learning to enhance MEMS IMU sensor performance.

In the initial phase, you will focus on software, simulation, and data-driven modeling using datasets from our core engineering group. You will train models to denoise signals, fuse data, and correct stochastic errors, with cross-border collaboration to