PhD position

AI Methods for Quantum Algorithms in Chemistry

• Design AI methods that assist in the selection, control, and assessment of quantum algorithms for quantum-chemical problems (electronic structure, reaction-energy profiles, spectral properties in catalysis and complex environments).
• Integrate AI models into hybrid quantum-classical workflows to identify subproblems suitable for quantum treatment, configure realistic quantum simulations, and analyze outputs.
• Leverage chemistry foundation models as feature extractors and priors to encode molecules, conformers, and reaction conditions into high-quality representations for downstream ML components interfaced with quantum algorithms.
• Investigate three key AI-for-quantum strategies:
  1. Learning-based active-space and embedding choices in multireference or fragment-based schemes (assigning parts of a system to quantum resources).
  2. Data-driven construction and adaptation of circuits for variational and sampling-based quantum diagonalization (e.g., SQD, SqDRIFT) on noisy hardware.
  3. ML-assisted parameter initialization and post-processing to accelerate convergence and sampling.
• Generate and curate synthetic training data from validated quantum-chemistry calculations and circuit simulations to address data sparsity in chemistry.
• Benchmark rigorously against established quantum-chemical baselines and analyze scaling and robustness.
  
  

• Enrollment eligibility for a PhD: a Master’s degree (or equivalent) in Computer Science, Electrical Engineering, Physics, Chemistry, Applied Mathematics, or a closely related field.
• Hands-on experience with quantum computing (e.g., Qiskit or other SDKs): circuit design/optimization; execution on simulators and/or real devices.
• Foundational knowledge in AI/ML (optimization, model training, regularization, metrics) and demonstrable practice applying ML to scientific or engineering problems.
• Strong Python skills and sound software practices (Linux, git/GitHub, testing, reproducibility).
• Familiarity with experiment orchestration, containers, and CI/CD.
• Clear communication in English and a structured, collaborative working style.

• Solid background in algorithm development and linear algebra.
• Experience with electronic- structure methods (e.g., DFT, MP2/CC, multireference approaches, embedding) or molecular spectroscopy.
• Exposure to variational and sampling-based quantum algorithms (e.g., VQE, SQD) and noise-aware circuit optimization.
• Experience with chemistry foundation models, graph neural networks, or transformers for molecular/reaction representations.
  
  

• Location: IBM Research Europe - Zurich, Switzerland; research visits at TU/e as appropriate.
• Duration: Typically, 4 years.
• Supervision: Joint-supervision by researchers from AI and Quantum team at IBM Research Europe - Zurich and TU/e.
  
  

Footer links