Embedded Edge AI Intern: Real-Time Anomaly Detection, Eindhoven

About Signify Through bold discovery and cutting‑edge innovation, we lead an industry that is vital for the future of our planet: lighting. Through our leadership in connected lighting and the Internet of Things, we’re breaking new ground in data analytics, AI, and smart solutions for homes, offices, cities, and beyond.

More About The Role Signify is one of the few companies in the world to achieve carbon neutrality and our next sustainability goals are even bolder: doubling our positive impact on the environment and society by 2025. As an embedded software intern, you will contribute to innovative projects focused on exploring how the latest advancements in Artificial Intelligence (AI), Machine Learning (ML), and Edge computing can be adopted to wireless embedded devices. These projects investigate how intelligent data processing and decision‑making can be brought closer to the source of data, optimizing models for resource‑constrained embedded devices, improving wireless bandwidth usage, and enhancing sensory detection algorithms. You’ll gain hands‑on experience with cutting‑edge technologies and help shape the future of smart, connected IoT lighting systems.

Topic Description Performance Evaluation and Application of Edge AI on Silicon Labs EFR32MG24 for Real‑Time Anomaly Detection. This project investigates the viability of deploying lightweight machine learning models on the EFR32MG24 (or a similar) microcontroller, focusing on its integrated AI/ML accelerator. The goal is to evaluate the chip’s computational performance, power efficiency, and real‑time inference capabilities for practical edge applications such as anomaly detection in sensor data.

What You’ll Do

Perform literature review on related works and evaluate a selected list of methodologies.

Select and optimize ML models for anomaly detection using embedded frameworks such as TensorFlow Lite for Microcontrollers.

Benchmark inference latency and memory usage of selected ML models; analyze the AI/ML hardware capabilities of the EFR32MG24 chip.

Profile energy consumption during inference with and without the hardware accelerator.

Develop a real‑world anomaly detection use case.

Propose an approach to integrate the solution into existing embedded data‑processing workflows.

Your Qualifications

Enrolled in an engineering degree in Embedded Systems, Software Engineering, Computer Science, or a related subject for the duration of the internship.

Passionate about software development.

Open to learning new skills and programming languages.

Experience managing code repositories using GIT.

Strong C programming skills and experience with at least one of C++, Java, C#, or Python.

Optional: prior experience in AI/ML related projects.

Available for 6–12 months, 32–40 hours per week, with a preferred start date of October 1.

What You’ll Get In Return

Continuous mentoring and feedback throughout the internship.

An internship allowance, including support for accommodation or public transport.

Paid leave equivalent to one day per month.

Access to a personalized learning platform and a wide range of development tools.

Discounts on products in the staff shop.

Opportunities for career development, including mentoring, coaching, and stretch assignments.

Participation in the Netherlands intern program, with events, training, and networking opportunities.

Notes

EU students: You must be registered as a student for the entire internship period; formal documentation may be requested.

Non‑EU students attending Dutch universities: You must be registered as a student, your university must sign the Nuffic Agreement, and the internship must be compulsory (receive ECTS). For non‑compulsory internships, a letter from the university detailing learning goals and tasks is required.

#J-18808-Ljbffr