yardenas/safe-sim2real
Sep 2025 – May 2026
2 Months active
Languages Used
YAMLPython
Technical Skills
Configuration ManagementPythonPython programmingalgorithm developmentmachine learningreinforcement learning
ManuelWendl
PROFILE
Manuelwendl
Contributed to the yardenas/safe-sim2real repository by delivering two targeted features over two months, focusing on experiment tracking and policy evaluation infrastructure. Implemented a configuration update in YAML to WandB logging, ensuring reliable experiment attribution during the transition from offline to online policy on rccar hardware. Later, refactored the backup action subsystem in ModelBasedEnv using Python, consolidating logic, introducing a dedicated inference function, and normalizing parameters to improve performance, stability, and maintainability. Leveraged skills in configuration management, algorithm development, and reinforcement learning to enhance code quality, support robust analytics, and lay groundwork for safer, more extensible policy evaluation workflows.
Overall Statistics
Feature vs Bugs
100%Features
Repository Contributions
4Total
Bugs
0
Commits
4
Features
2
Lines of code
44
Activity Months2
Your Network
7 peopleShared Repositories
7
Benjamin UngerMember
Benjamin UngerMember
Benjamin UngerMember
Benjamin UngerMember
Yarden AsMember
Yarden AsMember
YardenMember
Work History
May 2026
3 Commits • 1 Features
May 1, 2026May 2026 monthly summary for yardenas/safe-sim2real: Key engineering delivery focused on performance, stability, and maintainability of the backup action subsystem in ModelBasedEnv. Delivered a refactor of backup action calculation, backup policy, and action retrieval, consolidating logic, removing direct tanh usage, and introducing a new inference function and normalized parameters to improve numerical stability and readability. This work positions the system for easier future enhancements and safer policy evaluation under varying scenarios.
3 Commits • 1 Features
May 1, 2026May 2026 monthly summary for yardenas/safe-sim2real: Key engineering delivery focused on performance, stability, and maintainability of the backup action subsystem in ModelBasedEnv. Delivered a refactor of backup action calculation, backup policy, and action retrieval, consolidating logic, removing direct tanh usage, and introducing a new inference function and normalized parameters to improve numerical stability and readability. This work positions the system for easier future enhancements and safer policy evaluation under varying scenarios.
May 2026
September 2025
1 Commits • 1 Features
Sep 1, 2025September 2025 (Month: 2025-09) focused on enabling reliable experiment tracking during the offline-to-online policy transition on the rccar hardware within the yardenas/safe-sim2real project. Delivered a targeted configuration change to WandB logging to ensure experiments are recorded under the correct identifier, supporting accurate, centralized analytics and faster troubleshooting as we move toward online policy deployment.
September 2025
1 Commits • 1 Features
Sep 1, 2025September 2025 (Month: 2025-09) focused on enabling reliable experiment tracking during the offline-to-online policy transition on the rccar hardware within the yardenas/safe-sim2real project. Delivered a targeted configuration change to WandB logging to ensure experiments are recorded under the correct identifier, supporting accurate, centralized analytics and faster troubleshooting as we move toward online policy deployment.
Activity
Loading activity data...
Quality Metrics
Correctness90.0%
Maintainability90.0%
Architecture90.0%
Performance90.0%
AI Usage25.0%
Skills & Technologies
Programming Languages
PythonYAML
Technical Skills
Configuration ManagementPythonPython programmingalgorithm developmentmachine learningreinforcement learning
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline