I love all aspects of AI efficiency and am interested in exploring both its scientific foundations and system-level execution. On the scientific side, I focus on optimization algorithms, training dynamics, and sample efficiency to understand how models converge and generalize with minimal compute and data. On the engineering side, I am interested in both per-second and per-watt performance, aiming to make intelligence sustainable and efficient on everything from edge devices to massive computing clusters. Recently, I have also been focusing on recursive self-improvement, investigating methods to make agentic systems like autoresearch and deep research more efficient, both in reliably meeting their goals through harness engineering and in optimizing resource utilization by achieving results faster with less compute.

Papers

1. In-Context Learning Functions with Varying Number of Minima
   David Oniani, Yanshan Wang
   arXiv (2023)
2. Few-Shot Learning for Clinical Natural Language Processing Using Siamese Neural Networks: Algorithm Development and Validation Study
   David Oniani, Premkumar Chandrasekar, Sonish Sivarajkumar, Yanshan Wang
   JMIR AI (2023)
3. Large Language Models Vote: Prompting for Rare Disease Identification
   David Oniani, Jordan Hilsman, Hang Dong, Fengyi Gao, Shiven Verma, Yanshan Wang
   arXiv (2023)

I have been developing AI methods for various scientific disciplines, including healthcare and life sciences, chemistry, and, most recently, economics. The overarching goal is to leverage AI to advance scientific disciplines, uncover new phenomena, and solve both fundamental and applied problems across the natural and social sciences.

Papers

1. Generation-Augmented Retrieval for Product Recommendations
   David Oniani, Nathan Liu, Jin Shang, Alex Patry, Suju Rajan
   AMLC 2025 (2025)
2. A Scoping Review of Artificial Intelligence for Precision Nutrition
   Xizhi Wu, David Oniani, Zejia Shao, Paul Arciero, Sonish Sivarajkumar, Jordan Hilsman, Alex E Mohr, Stephanie Ibe, Minal Moharir, Li-Jia Li, Ramesh Jain, Jun Chen, Yanshan Wang
   Advances in Nutrition (2025)
3. Generative LLMs for Query-Product Relevance: Findings and Observations
   David Oniani, Amjad Jbara
   AMLC Workshop on Gen AI, Personalization Algorithms and Recommender Systems (2024)
4. Clinical Information Retrieval: A scoping review
   Sonish Sivarajkumar, Haneef Ahamed Mohammad, David Oniani, Kirk Roberts, William Hersh, Hongfang Liu, Daqing He, Shyam Visweswaran, Yanshan Wang
   Journal of Healthcare Informatics Research (2024)
5. Natural Language Processing for Digital Health in the Era of Large Language Models
   Abeed Sarker, Rui Zhang, Yanshan Wang, Yunyu Xiao, Sudeshna Das, Dalton Schutte, David Oniani, Qianqian Xie, Hua Xu
   Yearbook of Medical Informatics (2024)
6. Emerging Opportunities of Using Large Language Models for Translation Between Drug Molecules and Indications
   David Oniani, Jordan Hilsman, Chengxi Zang, Junmei Wang, Lianjin Cai, Jan Zawała, Yanshan Wang
   Scientific Reports (2024)
7. Generative AI enhanced with NCCN clinical practice guidelines for clinical decision support: A case study on bone cancer
   Yanshan Wang, Xizhi Wu, Luke Carlson, David Oniani
   ASCO 2024 (2024)
8. Extraction of Sleep Information from Clinical Notes of Alzheimer's Disease Patients Using Natural Language Processing
   Sonish Sivarajkumar, Thomas Yu CHow Tam, Haneef Ahamed Mohammad, Samual Viggiano, David Oniani, Shyam Visweswaran, Yanshan Wang
   JAMIA (2024)
9. Toward Improving Health Literacy in Patient Education Materials with Neural Machine Translation Models
   David Oniani, Sreekanth Sreekumar, Renuk DeAlmeida, Dinuk DeAlmeida, Vivian Hui, Young Ji Lee, Yiye Zhang, Leming Zhou, Yanshan Wang
   AMIA Informatics Summit (2023)
10. Comparisons of Graph Neural Networks on Cancer Classification Leveraging a Joint of Phenotypic and Genetic Features
    David Oniani, Chen Wang, Yiqing Zhao, Andrew Wen, Hongfang Liu, Feichen Shen
    arXiv (2021)
11. Leveraging a Joint of Phenotypic and Genetic Features on Cancer Patient Subgrouping
    David Oniani, Chen Wang, Yiqing Zhao, Andrew Wen, Hongfang Liu, Feichen Shen
    arXiv (2021)
12. Social and Behavioral Determinants of Health in the Era of Artificial Intelligence with Electronic Health Records: A Scoping Review
    Anusha Bompelli, Yanshan Wang, Ruyuan Wan, Esha Singh, Yuqi Zhou, Lin Xu, David Oniani, Bhavani Singh Agnikula Kshatriya, Joyce (Joy) E. Balls-Berry, Rui Zhang
    Health Data Science (2021)
13. Constructing Co-occurrence Network Embeddings to Assist Association Extraction for COVID-19 and Other Coronavirus Infectious Diseases
    David Oniani, Guoqian Jiang, Hongfang Liu, Feichen Shen
    JAMIA (2020)

I work toward AI systems that are safe, reliable, ethical, and beneficial to humanity. I care about alignment across the entire lifecycle, from behavioral guarantees and objective specification to post-deployment safety. The aim is to build trustworthy systems that adhere to human intent and operate responsibly in high-stakes and mission-critical environments.

Papers

1. Foundation metrics for evaluating effectiveness of healthcare conversations powered by generative AI
   Mahyar Abbasian, Elahe Khatibi, Iman Azimi, David Oniani, Zahra Shakeri Hossein Abad, Alexander Thieme, Ram Sriram, Zhongqi Yang, Yanshan Wang, Bryant Lin, Olivier Gevaert, Li-Jia Li, Ramesh Jain, Amir M. Rahmani
   npj Digital Medicine (2024)
2. Enhancing Large Language Models for Clinical Decision Support by Incorporating Clinical Practice Guidelines
   David Oniani, Xizhi Wu, Shyam Visweswaran, Sumit Kapoor, Shravan Kooragayalu, Katelyn Polanska, Yanshan Wang
   Human-Centred XAI: Enhancing AI Acceptability for Healthcare (IEEE ICHI Workshop) (2024)
3. Adopting and expanding ethical principles for generative artificial intelligence from military to healthcare
   David Oniani, Jordan Hilsman, Yifan Peng, Ronald K. Poropatich, Jeremy C. Pamplin, Gary L. Legault, Yanshan Wang
   npj Digital Medicine (2023)
4. A Qualitative Evaluation of Language Models on Automatic Question-Answering for COVID-19
   David Oniani, Yanshan Wang
   ACM-BCB (2020)

With autonomous agents now writing code, my interest centers on agentic software engineering and system design. I explore how we architect, verify, and orchestrate software when agents handle the implementation. This involves creating robust execution environments, feedback loops, and guardrails that ensure correctness and maintainability. The goal is to discover the system abstractions needed to reliably build and evolve software in an agent-driven world.

Papers

1. The ENACT network is acting on housing instability and the unhoused using the open health natural language processing toolkit
   Daniel R. Harris, Sunyang Fu, Andrew Wen, Alexandria Corbeau, Darren Henderson, Jordan Hilsman, David Oniani, Yanshan Wang
   JCTS (2024)
2. ReDWINE: A clinical datamart with text analytical capabilities to facilitate rehabilitation research
   David Oniani, Bambang Parmanto, Andi Saptono, Allyn Bove, Janet Freburger, Shyam Visweswaran, Jonathan Silverstein, Michael Becich, Anthony Delitto, Elizabeth Skidmore, Yanshan Wang
   International Journal of Medical Informatics (2023)

Papers

1. Setting Up Python Development Environment for Use in a Small Classroom
   Roman Yasinovskyy, Karina Hoff, David Oniani
   Midwest Instruction and Computing Symposium (2020)

Papers

1. Cosine Similarity and Its Applications in the Domains of Artificial Intelligence
   David Oniani
   (2020)
2. The Topology of Robotic Configuration and Motion Planning
   David Oniani
   (2019)
3. Type Inference Rules For Container Types in CCL
   David Oniani
   (2019)
4. Textual and Statistical Analysis of Russian IRA Facebook Advertisements
   David Oniani
   (2019)