Kaveh Taraghi Khah
🤖 Robot Learning | Researcher-in-Training 🎓 Incoming MSc (research-track) 💻 4+ yrs Software Engineering background
I bring 4+ years of applied systems and backend engineering experience into a growing focus on reinforcement learning and robot learning research.
Core interests:
- Reinforcement Learning (model-based & model-free)
- Robot Learning & Sim-to-Real Transfer
- Robotic Manipulation & Locomotion
- Deep Learning for Control
- Simulation Environments (MuJoCo, Isaac Lab, PyBullet)
- Robot Perception
Long-term goal: contributing to learning-based methods that let robots operate robustly in the real world.
- Reading and reproducing papers in RL / robot learning
- Reinforcement learning experiments in simulated robotic environments
- Small sim-to-real transfer experiments
- Building toward original research projects at the intersection of RL and robotics
Building foundational depth in:
- Deep Reinforcement Learning theory and implementation
- Robotics fundamentals (kinematics, dynamics, control)
- Simulation tooling for robot learning
- B.Sc. in Software Engineering
- Preparing for MSc (research-track) in ML/Robotics
- Open to a PhD path in Robot Learning, depending on how research develops during the MSc
4+ years in backend and distributed systems engineering, built alongside my undergraduate degree:
- Systems Programming (C++)
- Backend Engineering (Node.js, TypeScript, NestJS)
- Distributed Systems & Microservices
- DevOps & Infrastructure (Docker, Kubernetes, CI/CD)
- Observability & Performance Debugging (Prometheus, Grafana) This background gave me practical experience with large-scale, performance-aware system design, skills directly relevant to building and scaling RL training pipelines and real-time robotic control systems, which I'm now extending into formal research.
- Learn from data, not just rules
- Simulate first, deploy carefully
- Measure → Analyze → Improve
- Bridge systems engineering with learning-based robotics
To work on robots that learn, focusing on:
- Reinforcement learning for real-world robotic control
- Sim-to-real transfer
- Robust, data-efficient robot learning at scale
- Research: kaveh.taraghikhah@gmail.com
- Work: kaveh.tkh@gmail.com
- GitHub: https://github.com/kaveh-taraghikhah
- ORCID: https://orcid.org/0009-0000-3349-2716
⭐ Building toward a research career in robot learning