DC13 — Fleet-Level Workload Shaping with Live Grid Signals

Host: Technical University of Denmark (DTU), Lyngby, Denmark — Embedded Systems Engineering section (ESE), DTU Compute.
Work package: WP3 (Fleet Operations and Industrial Validation)
Duration and start: 36 months, full-time; start between February and September 2027 (latest start September 2027).

The project

In this project you build the brain of the fleet: the central engine that turns many separate sites into one carbon-efficient grid resource.

Reacting to live grid signals, the orchestrator makes two kinds of decision: workload shaping (where to place, pause, or move work) and flexibility dispatch (which sites should change their power using batteries, cooling, and adaptable compute). It must scale to hundreds of sites and act across timescales from sub-second to hourly.

What you will work on:

• Design a multi-agent orchestrator that manages 100 to 1000 sites.
• Implement multi-timescale control, from sub-second to hourly.
• Develop contextual bandits with bounded-regret guarantees.

What you will deliver:

• An orchestration prototype with measured decision latency and scalability across various fleet sizes.
• A counterfactual analysis toolkit that estimates the carbon savings of different orchestration policies.

Secondments

You will spend two research visits with partners in the network:

• Center Denmark (Denmark) — 3 months from around October 2027 (industry), with Tore Kjeld: study how renewable energy sources can be combined with datacenter operations.
• Technische Universität Berlin (TU Berlin) (Berlin, Germany) — 4 months from around January 2029 (academic), with Stefan Schulte: build the distributed meta-OS integration with the orchestrator and coordinate fleet-level security mechanisms.

The host

The Embedded Systems Engineering section at DTU Compute works on methods, tools, and architectures for dependable edge computing, edge AI, and low-energy computing.

Supervision

Main supervisor: Paul Pop (ORCID) — Professor of Cyber-Physical Systems and Head of ESE, DTU Compute; AEGIS Coordinator. He works on dependable edge computing and green computing, and coordinated the FORA MSCA network (2017–2021).

Co-supervisors:

• Tore Kjeld — Center Denmark
• Stefan Schulte (ORCID) — TU Berlin

Profile we are looking for

• Optimization, operations research, or reinforcement learning (multi-agent, bandits)
• Distributed systems
• MSc (completed or near completion) in computer science, applied mathematics, electrical engineering, or a related field
• Curiosity, initiative, and good spoken and written English

You must also meet the MSCA eligibility rules — no PhD yet, the mobility rule, and open to all nationalities.

How to apply

Official vacancy at the host institution: link coming soon.

Until then, apply through the central AEGIS application portal: AEGIS application portal (reference position DC13), and read what to include in your application.

Questions about this position? Contact the coordinator, Prof. Paul Pop, aegis@compute.dtu.dk.