2026 - Senior AI Researcher - Autonomous AI Agents for Telecommunication Networks - Permanent

Are you ready to define the next generation of global telecommunications? Join our world-class AI research team to lead the charge in developing true Autonomous AI Agents capable of managing and optimizing the future of zero-touch networks. ## What you'll do be instrumental in realizing ambitious goals like: - Zero-Touch Network Self-Healing: Autonomous observation, diagnosis, reasoning, and planning to achieve instant, intelligent network recovery. - Self-Optimization: Agents capable of observing network performance degradation, learning, and proactively implementing performance improvements without human intervention. While recent advancements in LLM-based Agentic AI have demonstrated remarkable decision-making potential, we aim to go further. This role involves pioneering work across: - Next-Generation Foundation Model Architectures beyond standard LLMs. - Robust and accurate Reasoning and Planning for mission critical network operations. - Agentic Reinforcement Learning (RL) - RL-based model fine-tuning for complex, sequential decision-making. - Agent Memory Management to enable robust, long-term learning and knowledge utilization. - Multi-agent learning, coordination, negotiation, and semantic communication in multi-agent systems. This is a unique opportunity to shape the strategic direction of AI-driven network autonomy, pushing the boundaries of what is technologically possible. We offer a platform for globally impactful work and a solid compensation package designed to attract and retain the best talent. Responsibilities As a Senior AI Researcher, you will be expected to provide technical leadership and execute high-impact research, including: - Continuously learn and assess the state-of-the-art of AI technologies, particularly those applicable to Autonomous Networks, and drive the adoption of relevant innovations. - Collaborate closely with internal domain experts to gain a deep understanding of core challenges, technological gaps, and strategic opportunities within the telecommunication domain. - Define the strategic research direction, develop cutting-edge solution designs, and provide technical guidance and mentorship to junior team members. - Contribute to the rapid and iterative development of validated software prototypes based on novel AI technologies, benchmarking alternative solutions rigorously. - Foster strong cross-functional relationships, working closely with Domain Experts, Data Engineers, other AI researchers, and product development teams to ensure successful research transfer. - Synthesize technical insights, proactively contribute to strategic planning to maintain Huawei's technological leadership, and propose innovative solutions and disruptive ideas for the future of Autonomous Telecommunication. ## What you'll bring PhD in Artificial Intelligence, Machine Learning, and/or Natural Language Processing. - 1+ years of industrial or post-doctoral research experience in Artificial Intelligence or Machine Learning, with a demonstrable history of high-impact research. - Strong hands-on experience with language model technology, especially with Large Language Models. Crucially, strong experience with foundation models applied to complex decision-making tasks is required. - Excellent mathematics and statistics background, with the ability to reason about algorithm properties and performance. - Strong programming skills in Python. - Practical, hands-on skills in the use of modern machine learning frameworks, such as PyTorch, TensorFlow, and Scikit-Learn. - Demonstrated problem-solving ability and proven track record of developing new algorithms and methodologies inspired by leading open source initiatives and academic research papers. ## Nice to have Strong domain knowledge of telecommunication networks (Telco networks) is a great advantage. - A strong publication track record in top-tier machine learning and computer science conferences and journals, including, but not limited to, ICML, ECML, NeurIPS (NIPS), AAAI, IJCAI, or IJCNN. …

See the listing for full requirements.