Embedded Software Engineer (m/f/d) GPU & edge AI Signal Processing

Thales; Ditzingen; Vor Ort; EUR 75.000 - 110.000; Vollzeit

Benefits dieser Stelle

Competitive compensation and benefits

Zusammenfassung

Thales in Ditzingen, Germany, is seeking an Embedded Software Engineer (m/f/d) to develop real-time sensor processing on embedded GPU compute modules.

You will implement high-throughput DSP algorithms, optimize pipelines for low latency, and work with CUDA, Linux, and edge‑AI SDKs on radar, lidar, imaging data.

Qualifikationen

• Completed technical studies in computer science, electrical or electronic engineering.
• Proficient in C++17+ on Linux with CUDA/GPU computing experience.
• Knowledge of real-time data streaming and heterogeneous memory architectures.
• Fundamentals of digital signal processing.
• Experience with embedded GPU / edge‑AI compute modules and SDK ecosystems.
• Experience with graph- or pipeline-based streaming frameworks for sensor or media processing.
• Python for prototyping, validation and tooling.
• Experience with high-rate sensor data (radar, lidar, imaging, SDR).
• Strong communication skills in small interdisciplinary teams.
• Self-directed, proactive with high ownership; travel ~15%.
• Very good German and English skills.

Aufgaben

• Implement and optimize high-throughput signal processing algorithms for GPU as parallel kernels.
• Optimize for throughput and latency using memory hierarchy, kernel fusion, asynchronous streams/graphs, mixed precision.
• Design zero-copy / DMA-based ingest paths from FPGA front-ends into GPU memory.
• Build deterministic, real-time capable streaming pipelines with buffering and precise time-stamping.
• Evaluate and adopt vendor SDKs, libraries and pipeline frameworks for edge AI inference.
• Maintain OS images, board support and deployment for embedded compute modules.
• Profile and benchmark; contribute unit/integration tests, CI/CD, documentation.

Kenntnisse

C++ (C++17+) on Linux

CUDA / GPU computing

Real-time streaming

Digital signal processing

Embedded GPU / edge‑AI SDKs

Graph / pipeline streaming frameworks

Python for prototyping

High-rate sensor data processing

Analytical / structured thinking

German & English proficiency

Travel willingness (~15%)

Ausbildung

Bachelor's degree in CS / Electrical Engineering

Tools

Linux

Vendor SDKs

Jobbeschreibung

Embedded Software Engineer (m/f/d) – GPU & edge AI Signal Processing

Location: Ditzingen, Germany

Position Overview

We are looking for an Embedded Software Engineer to bring real‑time sensor processing to life on embedded GPU compute modules. You will implement high‑throughput signal processing algorithms, build low‑latency streaming pipelines, and push the limits of edge hardware for data rates of several GB/s and strict latency budgets.

Responsibilities

• Implement and optimize high‑throughput signal processing algorithms for GPU (e.g., FFT, filtering, detection, estimation) as parallel kernels.
• Optimize for throughput and latency using memory hierarchy, kernel fusion, asynchronous streams/graphs, mixed precision.
• Design zero‑copy / DMA‑based ingest paths from FPGA front‑ends into GPU memory.
• Build deterministic, real‑time capable streaming pipelines including buffering, back‑pressure handling and precise time‑stamping.
• Evaluate and adopt vendor SDKs, libraries and pipeline frameworks for accelerated sensor processing and edge AI inference.
• Maintain OS images, board support and deployment for embedded compute modules.
• Profile and benchmark; validate against reference implementations; contribute unit/integration tests, CI/CD, documentation.

Qualifications

• Completed technical studies in computer science, electrical or electronic engineering.
• Strong modern C++ (C++17 or newer) on Linux.
• Hands‑on CUDA / GPU computing experience: kernel development, profiling, performance optimization.
• Understanding of real‑time data streaming and heterogeneous memory architectures.
• Fundamentals of digital signal processing.
• Experience with embedded GPU / edge‑AI compute modules and SDK ecosystems.
• Experience with graph‑ or pipeline‑based streaming frameworks for sensor or media processing.
• Python for prototyping, validation and tooling.
• Experience with high‑rate sensor data (radar, lidar, imaging, SDR).
• Strong communication skills in small interdisciplinary teams.
• Analytical and structured way of thinking.
• Self‑directed, proactive working style with high ownership.
• Quality awareness, especially in verification and interface definition.
• Hands‑on prototyping mindset – willingness to iterate fast.
• Willingness to travel for integration and measurement campaigns (~15%).
• Very good German and English skills.

Benefits

Competitive compensation and benefits package. Opportunities to work on cutting‑edge defense and security technologies.