Jira Metrics
Projects/Test AI Agents(TAA)

Throughput

How many issues were completed per week.

12

completed in range

9

by AI agents

Contributing tickets (12)

šŸ”
12
Key↕Summary↕Assignee↕Completedā–¼
TAA-18Model registration silently didn't happen on completed training jobs with register_model=trueSemenyuk Dmitriy7/7/2026
TAA-20Fail fast when a training config needs the solver but the job image doesn't have itOliverAI7/7/2026
TAA-19Training runs don't log the demonstration-imitation signal (bc_loss / bc_weight) — add the instrumentationOliverAI7/7/2026
TAA-16Autoplay orchestrator: serve the new 3900-input observation schema (blind_v8) alongside the current 1011-input oneOliverAI7/7/2026
TAA-13Wire parallel eval to MLflow and verify it matches serial on the real RL championSemenyuk Dmitriy7/4/2026
TAA-7Make the canonical lever-screen statistically valid (train-to-budget + N≄2000), forbid underpowered verdictsOliverAI7/4/2026
TAA-12Deterministic (argmax) eval arm logs NaN/empty win_rate instead of a numberOliverAI7/3/2026
TAA-11Rule-server loss-heuristic false-positives on draw-heavy-but-winnable positionsOliverAI7/3/2026
TAA-6Background solvable-seed collector (Azure CPU job)OliverAI7/3/2026
TAA-5TAA-2 follow-up: run the argmax-collapse E1-E7 investigation and deliver the findings doc (PR #34887 added only the tooling)Semenyuk Dmitriy7/1/2026
TAA-2Root-cause the argmax-collapse: why does greedy (deterministic) eval score ~0% when stochastic scores ~43%?OliverAI7/1/2026
TAA-1Orchestrator autoplay UI: always show the real/actual seed for RL agentsOliverAI7/1/2026
Showing 1–12 of 12
Page 1 of 1