What will you do as an Agentic & Generative Edge AI Optimization Engineer?
In this role, you will work on advanced AI systems that bring Generative AI capabilities to edge devices and embedded platforms. You will be part of an ambitious, multidisciplinary environment where AI research meets real-world applications.

Your focus will be on optimizing and deploying large language models and multimodal AI systems for resource-constrained environments, ensuring they run efficiently and reliably on specialized hardware accelerators.

Your responsibilities include:

Optimizing AI Models for Edge Deployment

Optimize large language models (LLMs) and multimodal models for efficient on-device deployment.
Apply advanced techniques such as quantization (8-bit, 4-bit, mixed precision), pruning, and distillation.
Adapt models for execution on embedded AI accelerators and NPU-based platforms.

Accelerating AI Inference

Investigate and implement system-level optimizations such as speculative decoding and efficient decoding algorithms.
Improve performance, latency, and memory usage of AI systems in edge environments.

Engineering Agentic AI Systems

Design and implement agentic AI capabilities, including tool orchestration and function calling.
Explore techniques for enabling tiny agents on edge devices using small language models.
Implement safe and reliable agent workflows, including guardrails and secure tool invocation.

Working with AI Inference and Deployment Frameworks

Deploy optimized models using frameworks such as Ollama, llama.cpp, ONNX Runtime, and TensorFlow Lite.
Ensure models run efficiently across different hardware platforms.

Benchmarking AI Systems

Develop benchmarking pipelines to measure performance, efficiency, and reliability of Generative and Agentic AI systems running on-device.

Developing Proof-of-Concepts

Build demonstrators and technology PoCs for edge AI use cases, such as intelligent sensing systems, industrial monitoring, or smart embedded applications.

From Research to Product

Translate state-of-the-art AI research into production-ready implementations.
Collaborate with multidisciplinary teams to integrate AI innovations into hardware and software platforms.

Solliciteren