About the role

In this role, you will help translate advanced AI research into production-ready solutions for edge environments. Your focus will be on optimizing large language models, improving system performance, and developing agentic AI capabilities that can run efficiently on resource-constrained hardware.

Main responsibilities

• Optimize LLMs and multimodal models for deployment on edge and embedded devices.
• Apply model optimization techniques such as quantization, pruning, and distillation to improve performance and efficiency.
• Improve inference performance through system-level optimizations and efficient decoding strategies.
• Develop and implement agentic AI capabilities, including tool orchestration and function calling.
• Design secure and reliable agent workflows, incorporating guardrails and safe tool invocation mechanisms.
• Deploy optimized models using inference engines and frameworks such as llama.cpp, ONNX Runtime, TFLite, and Ollama.
• Build benchmarking pipelines to evaluate the performance of generative and agentic AI systems on-device.
• Develop proofs of concept and demonstrators for edge AI use cases.
• Translate research innovations into production-ready implementations and collaborate with engineering teams to integrate them into products.

What you bring

• MSc, EngD, or PhD in Computer Science, AI, or a related technical field.
• 5+ years of experience in software or AI engineering with strong exposure to LLMs, VLMs, and performance optimization.
• Experience with model optimization techniques such as quantization, pruning, and efficient inference strategies.
• Strong experience with AI frameworks such as PyTorch or TensorFlow.
• Experience with agentic AI frameworks (e.g., LangChain or similar ecosystems).
• Understanding of safety and security mechanisms for AI agents, including guardrails and secure function calling.
• Experience with AI deployment toolchains and inference engines (e.g., CUDA, TensorRT, ONNX, TFLite).
• Experience working with embedded systems, NPUs, or edge AI hardware.
• Strong programming skills in Python, C/C++, and Linux environments.
• Familiarity with MLOps environments, build systems, and cross-compilation workflows is a plus.
• Strong communication skills and experience working in international and cross-functional teams.