I am driven by a fascination with aligning AI systems with human values. My experience in data analysis and modeling has laid a strong foundation for exploring the frontiers of safe and reliable AI. My research interests lie in mitigating risks surrounding "hallucination" and adversarial attacks, with a particular focus on red teaming and developing robust evaluation frameworks.
I believe in a future where AI empowers humanity, and I'm dedicated to contributing towards building that future responsibly
Publications
1. The Future Remains Unsupervised (2023) - Deep Learning Indaba
* A position paper advocating for an integrated approach to language learning, combining unsupervised learning and reinforcement learning.
2. Effective Web Scraping for Data Scientists (2023) - Data Science and Artificial Intelligence Conference Kabarak University
* A study into the art of web scraping
3. Enhancing HIV Testing Indicator Reporting (2024) - Data Science and Artificial Intelligence Conference Kabarak University
* Data Science Approach for Identifying PMTCT Reporting Discrepancies
Research Focus
My current research focus areas are:
- Alignment
- Red-teaming
- Mechanistic Interpretability
- "Hallucination" control
Find more of my research insights and thought pieces on: