Specializing in Reinforcement Learning, Computer Vision, and Advanced MLOps.
I am currently a Master's student in Computer Science (AI) at the University of Southern California (USC) , with a background from Sharif University of Technology.
My focus lies at the intersection of Reinforcement Learning, Robotics, and Large Scale MLOps. [cite_start]I have experience building autonomous agents for imperfect-information games and deploying scalable ML pipelines on the cloud.
- 🔭 Current Research: Adversarial Co-Evolution of RL and VLM/LLM Agents.
- 🌱 Learning: Advanced MLOps, ROC, and Control Theory.
- 👯 Looking to collaborate on: Robotics simulation and Medical Imaging.
- 💬 Ask me about: Deep Reinforcement Learning (PPO), Computer Vision, and MLOps pipelines.
Languages
AI & ML Frameworks
Data Science
MLOps & Cloud
Databases & Systems
| Project | Description | Tech Stack |
|---|---|---|
| Risk-Scaled Steering in MoE | Developed token-aware steering for MoE language models using 3D delta tensors to dynamically scale expert activations for improved safety at inference. | Python vLLM PyTorch HuggingFace |
| Linguistic-Agnostic SER | Probing framework for Speech Emotion Recognition transformers to evaluate paralinguistic and acoustic knowledge encoding across hidden layers. | Python PyTorch HuggingFace |
| Adversarial Co-Evolution | Framework training high-performance PPO agents against LLMs in card games using curriculum learning and knowledge distillation. | Python Ollama PPO |
| Multi-Modal Sentiment Classification | Tool for multi-modal sentiment analysis and time dynamics exploration within image-text conversations. | Python Pandas PyTorch |
