Scholar
Nawaf Alampara
Google Scholar ID: R7K3xFIAAAAJ
PhD Researcher, Friedrich Schiller University Jena
Citations & Impact
Citations
157
H-index
4
i10-index
3
Publications
13
Co-authors
9
list available
Publications
Resume (English only)
Academic Achievements
- 2025: Published a paper in Nature Chemistry on evaluating the chemical knowledge and reasoning abilities of large language models; 2024: Published a paper in Nature Computational Science on probing the limitations of multimodal language models for chemistry and materials research; 2024: Presented a paper at AI4Mat-Vienna on revealing Transformer models' limitations in capturing 3D geometric information crucial for materials modeling; 2025: Contributed to ChemPile, a 250GB diverse and curated dataset for chemical foundation models; 2025: Published a paper in Computational Materials Science on lessons from evaluating ML systems in materials science; 2024: Published a paper in Journal of Physics D: Applied Physics on the formation of an extended defect cluster in cuprous oxide.
Research Experience
- Nov 2023 - Present: PhD Researcher at Friedrich-Schiller-Universität Jena, Advisor: Dr. Kevin Maik Jablonka; Nov 2023 - Apr 2024: AI Research Contractor (Part-time) at Stability AI, Dataset curation | Benchmarking; Jun 2022 - Sept 2023: Principal Engineer at QpiVolta Technologies, Material simulation using geometric deep learning models | Software development; Jun 2021 - Jun 2022: Research Engineer at QpiAI Technologies, Real-time video analytics | Computer vision.
Education
- 2023—2026: Friedrich-Schiller-Universität Jena, Germany, PhD in Machine Learning for Science, Advisor: Dr. Kevin Maik Jablonka; 2018—2020: Indian Institute of Technology Bombay, India, MSc in Energy Science, Advisor: Prof. K R Balasubramaniam, Thesis: Defects and Dopants in Cu₂O - DFT study; 2015—2018: Birla Institute of Technology Mesra, India, BSc in Physics.
Background
- A second-year PhD student, working on building machine learning systems to speed up scientific research, especially projects that involve both research and tooling. Recently, analyzing general-purpose AI models/systems to understand their limitations in scientific applications and interpreting why they fail. The goal is to design AI systems that are not only impressive on benchmarks but truly impactful for advancing science and research.
Miscellany
- Interests include blogging, active on social media platforms such as Twitter, GitHub, and LinkedIn.
