PhD Candidate in Learning and Optimization Algorithms for Datacenter Fleets

Job summary

A PhD fellowship is available in the Embedded Systems Engineering section focused on developing algorithms and software for optimizing AI workloads across datacenter fleets to enhance carbon efficiency. This role is part of a larger doctoral network involving multiple European partners.

Qualifications

• A two-year master's degree (120 ECTS points) or equivalent academic level.
• Strong foundation in algorithms and data structures, along with solid programming skills.
• Background in machine learning, reinforcement learning, combinatorial optimization, operations research, or distributed systems.
• Curiosity, initiative, and the ability to manage problems independently.
• Proficiency in spoken and written English.

Responsibilities

• Design a multi-agent orchestrator for managing multiple datacenter sites.
• Implement control mechanisms for decision-making across various timescales.
• Develop learning-based decision methods with bounded-regret guarantees.
• Create a toolkit for counterfactual analysis to evaluate carbon savings from different policies.
• Publish research findings in international conferences and journals.
• Collaborate with academic and industrial partners and contribute to departmental service work, including teaching assistance.

Skills

• Strong programming and algorithmic skills.
• Experience in machine learning and optimization techniques.

Education

• A master's degree or equivalent in a relevant field.

Tools

• Software development tools and programming languages relevant to algorithm design and optimization.