About Me
Eric Modesitt
AI Researcher & Software Engineer
Academic Foundation
I earned both my Bachelor of Science and Master of Computer Science degrees from the University of Illinois Urbana-Champaign. During my time at UIUC, I developed a deep appreciation for efficient AI research, mainly due to my curiosity and ambition with regards to studying language and intelligence, combined with not having the same resources as frontier labs which produce state of the art results.
Industry Impact
I’ve been fortunate to work across startups and large companies. During school, I interned with Google Summer of Code, Mr. Cooper, and Capital One (which led to a full-time offer). I also contributed to student-led orgs like Disruption Lab and NeurotechX@UIUC. I’m currently a Full-time Software Engineer at Capital One. See more on my experience.
Research Focus
My research centers on a concept I call “learning efficiency” - this applies to both artificial intelligence systems and human learning processes.
AI Learning Efficiency
- Domain Adaptation: How can we efficiently specialize general AI models for specific fields like astronomy, medicine, or law?
- Dataset Curation: What’s the most cost-effective way to collect high-quality training data from noisy web sources?
- Large Language Models: How can we optimize these powerful models for specific tasks without extensive retraining?
My ORBIT methodology, published in Findings of ACL 2025 (paper), demonstrates how to curate massive, high-quality, domain-specific datasets from noisy web sources. This work reduced dataset curation costs while improving model performance across multiple domains.
Human Learning Efficiency
Beyond AI research, I’m fascinated by how humans can learn more effectively. This interest stems from my belief that understanding human learning can inform better AI training methods, and vice versa. I regularly write about deliberate practice, intuition building, and efficient learning strategies on my blog.
Mentorship & Collaboration
I’m fortunate to work with exceptional researchers who have shaped my approach to both research and career development:
Professor Chengxiang Zhai (Current Research Mentor)
Currently collaborating with Professor Zhai, a distinguished researcher in information retrieval and text mining at UIUC. Under his guidance, I’m exploring advanced techniques in natural language processing and information retrieval, particularly focusing on how to make AI systems more effective at understanding and processing domain-specific information.
Professor Volodymyr Kindratenko (Former Research Mentor)
Previously worked with Professor Kindratenko, an expert in high-performance computing and AI acceleration at NCSA. This collaboration provided me with deep insights into the computational aspects of AI research and the importance of efficient implementation.
Technical Expertise
Programming Languages: Python, Java, JavaScript, SQL
AI/ML Frameworks: PyTorch, JAX, TensorFlow, Hugging Face Transformers
Data Processing: Pandas, NumPy, Apache Spark
Cloud Platforms: AWS, Google Cloud Platform
Development Tools: Git, Docker, Kubernetes
Personal Philosophy
I believe that the most meaningful progress in AI comes from:
- Bridging Theory and Practice: Academic research must be grounded in real-world applications
- Efficiency Over Scale: Smart methods often outperform brute-force approaches
- Human-Centered AI: Technology should augment human capabilities, not replace human judgment
- Continuous Learning: Both AI systems and their creators must be designed for continuous improvement
Beyond Work
When not coding or researching, I’m often thinking about learning efficiency in other contexts:
- How can we teach complex concepts more effectively?
- What makes some people naturally better learners?
- How can we apply AI insights to improve human education?
I also enjoy exploring the intersection of technology and society, particularly how AI advances can be made more accessible and beneficial to broader communities.
Hobbies
- To be updated!
Let’s Connect
I’m always interested in discussing:
- Research Collaborations: Particularly in domain adaptation, dataset curation, or learning efficiency
- Industry Applications: How to bridge academic research and practical implementation
- Career Advice: Navigating between academia and industry
- Learning Strategies: Both for AI systems and human learners
“The goal is not just to make AI more intelligent, but to make the entire learning process—for both machines and humans—more efficient and effective.”