Ava Kouhana
MS Student — Stanford ICME · Computational & Mathematical Engineering
Ava Kouhana completed her ICME master's at Stanford, graduated in June 2026. At Stanford, she researched Reinforcement Learning and World Models in Professor Gordon Wetzstein's lab. Outside from the lab, she decided to bridge the gap between her undergrad in Applied Maths, and a Computer Science degree, and took challenging classes : Operating System Design and Implementation (CS140E), Parallel Computing (CME 213), and also was a TA for the Machine Learning class (CS229). Prior to Stanford, spent six months at Harvard with Dr. Mengyu Wang on image segmentation and vision-language models for medical imaging fairness, and another six months with Dr. Craig Levin applying latent diffusion models to PET image reconstruction.
Ava is looking for her next opportunity in industry.
Email
LinkedIn
GitHub
Scholar
Publications
Projects
CoverAgent — Multi-Agent Collision in Responsibility Attribution
Studying whether LLM agents preserve accurate responsibility attribution when a peer agent pressures them to obscure its own role in an incident. The harness probes how human-facing reports change under multi-agent peer pressure.
Residual RL with Uncertainty for Dexterous Manipulation
Analyzed interaction between imitation learning and residual RL, showing that increased data can degrade performance.
Blog
Visit the blogCoverAgent: Responsibility Attribution Under Pressure to Conceal
A behavioral evaluation for whether AI assistants preserve accurate responsibility attribution when another agent pressures them to conceal fault.
PiDMASound: Bare-Metal Audio on a Raspberry Pi
Turning a single 1-bit GPIO pin into analog audio on bare metal — sigma-delta modulation, DMA, PWM, and interrupts.
Grants & Awards
- Recipient of 2024 IEEE Nuclear Science, Medical Imaging Trainee Grant (NSS MIC)
- 2nd place Outstanding Poster at the 20th Anniversary of ICME Stanford Research Symposium