I’m an AI researcher with a Ph.D. in Computer Science (minor: Statistics) from Iowa State University, and a current AI Safety Research Fellow at Algoverse.
My path into research started with a practical question: how do we build systems that behave reliably when the real world doesn’t cooperate? That question pulled me from software engineering at Comcast into a PhD focused on safety-critical control for smart energy infrastructure — and eventually toward the broader challenge of making AI systems trustworthy at scale.
I work at the intersection of deep reinforcement learning, AI safety, and physical systems — building agents that not only perform well, but can be trusted in high-stakes environments.
Current Focus
AI Safety Research Fellow @ Algoverse (Jan 2026 – present)
Designing and evaluating methods for agentic AI safety — including behavioral evaluations for reward hacking and deceptive alignment, adversarial robustness frameworks for LLM-based agents, and scalable oversight mechanisms. This builds directly on my doctoral work on physics-informed, safety-critical reinforcement learning.
Research
My doctoral work centered on physics-informed deep reinforcement learning for smart grid control: embedding domain knowledge, safety constraints, and uncertainty directly into the learning process so agents make reliable decisions under distribution shifts and partial observability.
Over time, “how do we keep agents safe?” scaled beyond energy systems. I now focus on AI safety and alignment more broadly:
🔬
Agentic Evaluations
Designing behavioral and mechanistic tests to detect reward hacking, deceptive alignment, and obfuscated internal states in LLM-based agents.
🛡️
Adversarial Robustness
Understanding and hardening AI systems against distributional attacks, prompt injection, and specification gaming in both classical and foundation model pipelines.
🔭
Scalable Oversight
Exploring how humans and increasingly capable AI systems can collaborate to maintain meaningful supervision at scale.
⚡
Safe RL for CPS
Physics-informed DRL with Bayesian uncertainty quantification, constraint satisfaction, and transfer learning for smart energy infrastructure.
🤖
LLM-Integrated Agents
Frameworks connecting perception, planning, and language reasoning for autonomous systems — bridging interpretable decision-making with low-level control.
👁️
Autonomous Perception
Vision-based perception (detection, segmentation, sensor fusion) integrated with learning-based control and trajectory planning.
For a full account of projects and publications, see the Research and Projects pages.
Technical Skills
AI / Machine Learning
Deep Reinforcement Learning
Physics-Informed Neural Networks
Bayesian Deep Learning
Uncertainty Quantification
Transfer Learning
Multi-Agent Systems
AI Safety & Alignment
Adversarial Robustness
Agentic Evaluations
Scalable Oversight
Reward Hacking Detection
Deceptive Alignment
Large Language Models & Agents
LangChain / LangGraph
RAG Pipelines
Tool-Use & Memory
Prompt Engineering
OpenAI / Anthropic APIs
HuggingFace Transformers
Languages & Frameworks
Python
PyTorch
TensorFlow
R
OpenAI Gym / Stable Baselines3
CARLA Simulator
OpenCV
scikit-learn
SQL
Git / Docker
Education & Experience
Minor in Statistics. Dissertation: "Safe and Robust Deep Reinforcement Learning for Constrained Control in DER-Integrated Smart Energy Grids." Advised by Dr. Christopher Quinn and Dr. Ravikumar Gelli.
Coursework spanned deep learning, NLP, computer vision, AI for cybersecurity, statistical theory, empirical methods, and experimental design.
Deep RL
Safety-Critical Control
Uncertainty Quantification
Bayesian ML
Smart Grids
Developing evaluation frameworks and robustness methods for agentic AI safety, including behavioral tests for reward hacking, deceptive alignment, and scalable human-AI oversight.
AI Safety
Agentic Evals
Adversarial Robustness
Six years supporting 300+ students across software development, database systems, and CS fundamentals. Held office hours, designed assignments, and mentored undergraduate research projects.
Software Engineering
Database Systems
CS Fundamentals
Worked in industry on software engineering before pivoting to research — motivated by the question of how to build systems that behave reliably when the real world doesn't cooperate.
Software Engineering
Systems
At a Glance
6+
Years Research
300+
Students Taught
10+
Publications
3
Research Areas
Beyond Research
I try to give back what I learn. I write on AI safety, reinforcement learning, and LLMs on Substack, and produce video explainers and walkthroughs on YouTube. Outside of work, I love cooking and ice skating 🛼.
View full CV → See my Research →
