I'm a PhD candidate in Computer Science at Colorado State University, advised by Prof. Nathaniel Blanchard. My research focuses on using NLP to understand how people think and collaborate — from detecting cognitive states in dialogue to building AI agents that intervene in group reasoning.
I work on DARPA and NSF-funded projects studying multi-party dialogue, collaborative problem solving, and the computational modeling of common ground. I'm also an ML Engineer at HVS, where I deploy production RAG systems.
How do groups of people build shared understanding, and how can AI help when that process breaks down?
My research sits at the intersection of natural language processing, cognitive science, and education. I study the computational structure of collaborative dialogue: how beliefs propagate through conversation, how reasoning chains form and fracture, and how cognitive states like confusion or familiarity surface in language and speech.
A central thread of my work is epistemic modeling, building systems that track what individuals know, believe, and assume within a group. This includes extracting propositional content from naturalistic speech, detecting collaborative problem-solving behaviors in real time, and designing AI agents that intervene meaningfully in high-stakes group tasks.
Looking ahead, I am interested in how large language models can be grounded in the messy, incremental nature of real human dialogue, and what it takes to make them trustworthy partners in collaborative reasoning rather than passive responders.
Minor in Mathematics & Philosophy
Minor in Mathematics & Philosophy
- Incoming intern on the AI2 team, focusing on model inference optimizations
- Engineered an end-to-end RAG pipeline using Llama 3.1 and FastAPI for on-premise property evaluation reports
- Integrated Python AI services with existing C# web infrastructure
- Collaborated with domain experts to iterate on retrieval accuracy and output quality
- Real-time propositional extraction from multi-party dialogues to model group common ground
- Framework for tracing deliberation chains and causal links in collaborative reasoning
- Human-AI conversational agent that intervenes to promote reflective reasoning — demoed live for DARPA
- Cognitive-affective state detection from speech using acoustic features and transformer embeddings
- Interpretable ML using VR eye-tracking features to detect the feeling of familiarity
- Built ML models for bird call identification to assist endangered species monitoring
- Vulnerability scanning on critical web applications using Burp Suite Enterprise; initiated remediation campus-wide
- Designed a wire identification system in C for aerospace manufacturing using I2C and SPI protocols
- Led lectures and labs for Distributed Systems, Big Data, Data Structures & Algorithms, and ML Foundations