Master's Thesis — Next-Generation Query and Workflow Processing across Heterogeneous Engines and Hardware

About the Role

Modern data platforms increasingly execute complex analytical and AI-driven workloads across heterogeneous engines and hardware. Queries and workflows may span SQL engines, vector databases, streaming systems, ML pipelines, and specialized accelerators. Efficiently routing, scheduling, and orchestrating these workloads across engines and resources has become a central systems challenge.

At IBM Research Europe – Zurich, we are building the next generation of intelligent data platforms that leverage analytical models, machine learning, and systems optimization to make better decisions about where, when, and how workloads should execute across heterogeneous software and hardware stacks. We are looking for Master’s students interested in solving challenging research problems at the intersection of:

• heterogeneous query and workflow processing
• AI-driven systems optimization
• database systems and distributed systems
• hardware-aware query processing
• machine learning for systems

Your work will combine systems research with machine learning and optimization, contributing to the design of next-generation heterogeneous data platforms that maximize performance, scalability, and cost efficiency.

Example Research Topics

• Opportunity-aware query processing
  Design intelligent systems that discover and exploit reusable execution opportunities across workloads, enabling query processing systems to reuse execution state, reduce redundant work, and improve cost-performance across heterogeneous engines.
• Workload shaping for heterogeneous systems
  Develop techniques that transform incoming workloads before execution by batching, grouping, reordering, or restructuring queries to improve throughput, latency, and resource efficiency.
• AI-native Query Control Planes
  Design the next generation of query control planes that continuously manage admission, scheduling, routing, execution state, and runtime adaptation across heterogeneous engines and hardware.
• Hardware-aware query optimization
  Build analytical or ML-based models that reason about CPUs, GPUs, accelerators, memory hierarchies, and data movement to optimize query placement and execution across heterogeneous computing environments.
• Cross-engine query and workflow optimization
  Represent and optimize complex workflows spanning SQL engines, vector databases, streaming systems, and machine learning pipelines, enabling efficient execution across heterogeneous software stacks.

What We Are Looking For

We seek highly motivated students interested in systems research for large-scale data platforms. You enjoy solving challenging systems problems that combine database systems, distributed systems, machine learning, and performance optimization. You are excited by understanding how software, query processing, and hardware interact, and enjoy reasoning about trade-offs across the entire systems stack—from query plans and execution engines to CPUs, GPUs, and emerging accelerators.

Minimum Qualifications

• Bachelor’s degree in Computer Science, Data Engineering, Systems, or a related field
• Strong programming skills
• Solid foundations in database systems, distributed systems, or systems optimization
• Experience with Python or similar programming languages
• Strong analytical and problem-solving skills
• Fluent English communication skills

Preferred Qualifications

• Experience with query processing or database internals
• Familiarity with query optimization and cost models
• Knowledge of modern data formats and interfaces (e.g., Parquet, Arrow, Iceberg, Substrait)
• Experience with distributed data processing frameworks
• Interest in machine learning for systems or AI-driven systems optimization
• Experience with GPU programming or heterogeneous computing is a plus

Diversity & Work Environment

You will join an open, multicultural research environment that values different perspectives and supports flexible working arrangements. Our goal is to help all genders and backgrounds thrive professionally while maintaining a healthy work–life balance.

How to Apply

Interested candidates please submit your application through the link below. Apply

Interview Process

After the initial screening based on the uploaded documentation, identified candidates will be contacted for a first technical discussion on their experience, background, and motivations, followed by coding interview and an AI/ML interview. Further selection steps might be added based on candidates skills and project needs.

If you have any question related to this position, please contact Prof. Dr. Bert Offrein, Manager Co-packaged Optics at ofb@zurich.ibm.com or tel. +41 44 724 8572. If you have any question related to this position, please contact Dr. Cezar Zota, zot@zurich.ibm.com.