Lead Time
Days from issue creation until it is resolved.
How long tickets actually take to finishactive · business hrs ÷ pauses
2.7d
Avg
3.1d
Active Time (÷ pauses)
3.7d
P75
4.3d
P85
4.6d
P90
Wall-clock (created → done)raw / Jira-flow · calendar days
4d
Avg
3.1d
Calendar Time (raw)
5d
P75
6.8d
P90
12
issues in range
Weekly Median trend
Contributing tickets (12)
🔍
12
| Key↕ | Summary↕ | Assignee↕ | Lead (active)▼ | Completed↕ |
|---|---|---|---|---|
| TAA-1 | Orchestrator autoplay UI: always show the real/actual seed for RL agents | OliverAI | 5d | 7/1/2026 |
| TAA-2 | Root-cause the argmax-collapse: why does greedy (deterministic) eval score ~0% when stochastic scores ~43%? | OliverAI | 4.6d(+0.4d paused) | 7/1/2026 |
| TAA-7 | Make the canonical lever-screen statistically valid (train-to-budget + N≥2000), forbid underpowered verdicts | OliverAI | 4.1d | 7/4/2026 |
| TAA-13 | Wire parallel eval to MLflow and verify it matches serial on the real RL champion | Semenyuk Dmitriy | 3.6d(+0.5d paused) | 7/4/2026 |
| TAA-12 | Deterministic (argmax) eval arm logs NaN/empty win_rate instead of a number | OliverAI | 3.1d | 7/3/2026 |
| TAA-11 | Rule-server loss-heuristic false-positives on draw-heavy-but-winnable positions | OliverAI | 3.1d | 7/3/2026 |
| TAA-6 | Background solvable-seed collector (Azure CPU job) | OliverAI | 3.1d | 7/3/2026 |
| TAA-18 | Model registration silently didn't happen on completed training jobs with register_model=true | Semenyuk Dmitriy | 1.6d | 7/7/2026 |
| TAA-20 | Fail fast when a training config needs the solver but the job image doesn't have it | OliverAI | 1.6d | 7/7/2026 |
| TAA-19 | Training runs don't log the demonstration-imitation signal (bc_loss / bc_weight) — add the instrumentation | OliverAI | 1.5d | 7/7/2026 |
| TAA-16 | Autoplay orchestrator: serve the new 3900-input observation schema (blind_v8) alongside the current 1011-input one | OliverAI | 1.5d | 7/7/2026 |
| TAA-5 | TAA-2 follow-up: run the argmax-collapse E1-E7 investigation and deliver the findings doc (PR #34887 added only the tooling) | Semenyuk Dmitriy | 0.1d(+2.6d paused) | 7/1/2026 |
Showing 1–12 of 12
Page 1 of 1