Engineer at Google, working on AI infrastructure for large-scale training. Independently, I research reasoning, multilingual safety evals, and interpretability.
Experience
Software Engineer
Keeping one of the world's largest GPU fleets upgraded, validated, and ready to train.
Member of Technical Staff
First engineer, building agents that read, retrieve, and cite.
Software Engineering Intern
Synthetic reasoning-chain data for agents in Google Agentspace.
Software Engineering Intern
LLM-based structured extraction from support conversations.
Machine Learning Research Assistant
Benchmarking how humans and AI compare on audio question answering.
Machine Learning Research Assistant
Studying what drives cross-lingual transfer in multilingual models.
Publications
Cross-Lingual Alignment for Decoder-Only Models using MoE Routers
L. Bandarkar, C. Peng, Ahmed Haj Ahmed, A. Khandelwal, N. Peng.
ADAGE: A Language-Agnostic Pipeline for Analogical Reasoning Evaluation
Ahmed Haj Ahmed, A. Grissom II.
AUDITA: A New Dataset to Audit Humans vs. AI Skill at Audio QA
T. Kabir, D. Kurdydyk, A. Palnitkar, L. Dorn, Ahmed Haj Ahmed, J. L. Boyd-Graber.
Disentangling Linguistic Relatedness from Task Alignment in Cross-Lingual Transfer
Ahmed Haj Ahmed, R. Zhang, A. Grissom II.
CuLEmo: Cultural Lenses on Emotion - Benchmarking LLMs for Cross-Cultural Emotion Understanding
T. D. Belay, Ahmed Haj Ahmed, A. Grissom II, et al.
Navigating Dialectal Bias and Ethical Complexities in Levantine Arabic Hate Speech Detection
Ahmed Haj Ahmed, R.-J. Yew, X. Minocher, S. Venkatasubramanian.