Spaces:
Sleeping
Sleeping
| spec_version: 1 | |
| name: engineer-manager | |
| type: space | |
| runtime: fastapi | |
| app: server.app:app | |
| port: 8000 | |
| version: 0.1.0 | |
| entry_point: server.engineer_manager_environment:EngineerManagerEnvironment | |
| api: | |
| base_url: / | |
| endpoints: | |
| reset: | |
| method: POST | |
| path: /reset | |
| step: | |
| method: POST | |
| path: /step | |
| state: | |
| method: GET | |
| path: /state | |
| tasks: | |
| method: GET | |
| path: /tasks | |
| grader: | |
| method: POST | |
| path: /grader | |
| tasks: | |
| - id: engineer_manager_task_0 | |
| task_id: quiet-morning | |
| name: quiet-morning | |
| difficulty: easy | |
| description: High-noise morning where the agent should mute comms early and protect an uninterrupted work block. | |
| max_steps: 32 | |
| reset_params: | |
| task_id: 0 | |
| action_schema: | |
| target_slot: integer slot index within the workday | |
| operation: 0=idle, 1=schedule work, 2=reschedule meeting, 3=toggle mute comms | |
| task_id: quiet-morning | |
| grader: graders:grade_task_0 | |
| graders: | |
| - graders:grade_task_0 | |
| reward_range: | |
| - 0.0 | |
| - 1.0 | |
| - id: engineer_manager_task_1 | |
| task_id: meeting-surgery | |
| name: meeting-surgery | |
| difficulty: medium | |
| description: Fragmented calendar where selective meeting moves should improve flow. | |
| max_steps: 32 | |
| reset_params: | |
| task_id: 1 | |
| action_schema: | |
| target_slot: integer slot index within the workday | |
| operation: 0=idle, 1=schedule work, 2=reschedule meeting, 3=toggle mute comms | |
| task_id: meeting-surgery | |
| grader: graders:grade_task_1 | |
| graders: | |
| - graders:grade_task_1 | |
| reward_range: | |
| - 0.0 | |
| - 1.0 | |
| - id: engineer_manager_task_2 | |
| task_id: delivery-triage | |
| name: delivery-triage | |
| difficulty: hard | |
| description: Constrained delivery day with hidden task complexity and tighter tradeoffs. | |
| max_steps: 32 | |
| reset_params: | |
| task_id: 2 | |
| action_schema: | |
| target_slot: integer slot index within the workday | |
| operation: 0=idle, 1=schedule work, 2=reschedule meeting, 3=toggle mute comms | |
| task_id: delivery-triage | |
| grader: graders:grade_task_2 | |
| graders: | |
| - graders:grade_task_2 | |
| reward_range: | |
| - 0.0 | |
| - 1.0 | |
| graders: | |
| - graders:grade_task_0 | |
| - graders:grade_task_1 | |
| - graders:grade_task_2 | |