Jammy2211 · Oct 23, 2026 · Oct 17, 2026 · Oct 17, 2026 · Oct 17, 2026 · Oct 23, 2026
diff --git a/campaigns.yaml b/campaigns.yaml
diff --git a/checkins/2026-10-17.md b/checkins/2026-10-17.md
@@ -88,3 +88,13 @@ and independent; merge #425 after #636.
 - `/mnt/ral/jnightin/scratch/fft_levers_stack.tgz`, `fft_levers_jobs/`, `fft_levers_results/` (results JSON is committed in #433)
 - `/mnt/ral/jnightin/profiling/autolens_profiling_wt/ps-image-admission_stale_*`
 - Worker private clones under `profiling/` and `inference/` are kept for reuse.
+
+## Post-/prm
+
+The human ran `/prm` on 2026-10-17 after this review; every PR above merged.
+
+- **Merged:** PyAutoArray [#636](https://github.com/PyAutoLabs/PyAutoArray/pull/636) and [#637](https://github.com/PyAutoLabs/PyAutoArray/pull/637) (both pending-release); autolens_profiling [#425](https://github.com/PyAutoLabs/autolens_profiling/pull/425), [#432](https://github.com/PyAutoLabs/autolens_profiling/pull/432), [#431](https://github.com/PyAutoLabs/autolens_profiling/pull/431), [#433](https://github.com/PyAutoLabs/autolens_profiling/pull/433), [#434](https://github.com/PyAutoLabs/autolens_profiling/pull/434), [#435](https://github.com/PyAutoLabs/autolens_profiling/pull/435) (main now c40e33c8, the merge of #435); autolens_inference [#29](https://github.com/PyAutoLabs/autolens_inference/pull/29); PyAutoPulse [#66](https://github.com/PyAutoLabs/PyAutoPulse/pull/66) and [#67](https://github.com/PyAutoLabs/PyAutoPulse/pull/67).
+- **Issues closed:** PyAutoArray #635, #629; autolens_profiling #427, #428, #429, #430, #368, #426; autolens_inference #25.
+- **Mind completion records:** [`direct-conv-symmetric`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/direct-conv-symmetric.md), [`delaunaynn-wide-row-overhead`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/delaunaynn-wide-row-overhead.md), [`ps-source-gradient-admission`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/ps-source-gradient-admission.md), [`imaging-production-over-sampling`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/imaging-production-over-sampling.md), [`point-source-image-gpu-admission`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/point-source-image-gpu-admission.md), [`ral-route-constants-fft-levers`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/ral-route-constants-fft-levers.md), [`interferometer-streaming-memo-after`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/interferometer-streaming-memo-after.md).
+- **Owed:** shadow row in Mind `autonomy_log` for the Aris L2 work (glance tier; human to confirm merged-unchanged).
+- Library speed-ups (#636, #637) are still unreleased; the figures above remain PR-time measurements until a release carries them. This section links close-outs and does not count as a check-in.
diff --git a/dashboard.html b/dashboard.html
diff --git a/dashboard.md b/dashboard.md
diff --git a/receipts/fit.json b/receipts/fit.json
@@ -13,5 +13,5 @@
   "record_count": 31,
   "errors": [],
   "cached_from": null,
-  "refreshed_at": "2026-10-23T17:38:22Z"
+  "refreshed_at": "2026-10-23T21:50:10Z"
 }
diff --git a/receipts/lens-walltime.json b/receipts/lens-walltime.json
@@ -13,5 +13,5 @@
   "record_count": 56,
   "errors": [],
   "cached_from": null,
-  "refreshed_at": "2026-10-23T17:38:22Z"
+  "refreshed_at": "2026-10-23T21:50:10Z"
 }
diff --git a/receipts/lens.json b/receipts/lens.json
@@ -3,15 +3,15 @@
   "repo": "autolens_profiling",
   "github": "PyAutoLabs/autolens_profiling",
   "branch": "main",
-  "resolved_commit": "da54dc30ff83c7682527984fa0e8a710a70a9927",
+  "resolved_commit": "aeef27565b64520ae01faf6200286aee842ace6b",
   "summary_path": "dashboard/catalogue.json",
   "source": "remote",
-  "fetched_at": "2026-10-23T17:38:00Z",
+  "fetched_at": "2026-10-23T21:50:09Z",
   "outcome": "ok",
   "schema": "profiling-summary",
   "version": 2,
   "record_count": 112,
   "errors": [],
   "cached_from": null,
-  "refreshed_at": "2026-10-23T17:38:21Z"
+  "refreshed_at": "2026-10-23T21:50:09Z"
 }
diff --git a/snapshots/lens.json b/snapshots/lens.json
@@ -3,8 +3,8 @@
   "repo": "autolens_profiling",
   "github": "PyAutoLabs/autolens_profiling",
   "summary_path": "dashboard/catalogue.json",
-  "commit": "da54dc30ff83c7682527984fa0e8a710a70a9927",
-  "fetched_at": "2026-10-23T17:38:00Z",
+  "commit": "aeef27565b64520ae01faf6200286aee842ace6b",
+  "fetched_at": "2026-10-23T21:50:09Z",
   "summary": {
     "archive_record_count": 21002,
     "comparison_policy": {

diff --git a/tasks/delaunay_nn_cpu_imaging.md b/tasks/delaunay_nn_cpu_imaging.md
@@ -58,3 +58,9 @@ Per the human decision of 2026-10-17 these stay as research:
 - Campaign created parked (human decision 2026-10-17, chat). #637 recorded at PR-open; it is file-
   disjoint from PyAutoArray#636 and independent of it.
 - [Full dated review](../checkins/2026-10-17.md).
+
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [PyAutoArray#637](https://github.com/PyAutoLabs/PyAutoArray/pull/637) (pending-release).
+- **Issues closed:** PyAutoArray#629.
+- **Mind record:** [`delaunaynn-wide-row-overhead`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/delaunaynn-wide-row-overhead.md).
diff --git a/tasks/hst_mge_cpu_pilot.md b/tasks/hst_mge_cpu_pilot.md
@@ -211,3 +211,10 @@ Pulse-side preparation under `--auto`, leaving the independent correctness audit
 current session. No witness or profiling execution, campaign freeze or acceptance
 was authorized. This focused preparation does not advance the all-campaign
 `last_checkin` or measurement dates.
+
+### Post-/prm 2026-10-17 — qualification design merged
+
+- autolens_profiling#418 merged 2026-10-17 under human `/prm` after an independent review (verdict FINDINGS, none blocking a doc-only merge; review posted on the PR). Issue #417 closed. Mind record: [complete/2026/10/hst-mge-pilot-qualification.md](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/hst-mge-pilot-qualification.md).
+- The design (`scripts/misc/jax_compile/QUALIFICATION.md`) derives what the pilot computes, proposes an independent operator + scipy NNLS witness, lists the identities to capture and gives candidate tolerances marked unaccepted. No contradiction with the #416 runner.
+- Carried, not fixed: no background-load method proposed; the 1e-8 relative image L2 budget is likely unattainable against the packaged solver's inactive-column flux allowance, so synthetic-harness calibration must be allowed to revise tolerances; correctness-process feasibility is deferred, so the 32-process / 16 CPU-hour budget stays unfinalized. Witness placement in autolens_workspace_test with a project adapter needs the human's confirmation.
+- No collection, campaign freeze or acceptance authorized.
diff --git a/tasks/imaging_production_over_sampling.md b/tasks/imaging_production_over_sampling.md
@@ -38,6 +38,13 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Phase 2** (A100 production re-measure, `--over-sampling production --instrument euclid|hst`): **needs-decision** — human undecided 2026-10-17 (chat). Requires #431 merged and separate compute authorization.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#431](https://github.com/PyAutoLabs/autolens_profiling/pull/431).
+- **Issues closed:** autolens_profiling#430.
+- **Mind record:** [`imaging-production-over-sampling`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/imaging-production-over-sampling.md).
+- Phase 2 (A100 production re-measure) still needs a human decision and compute authorization.
+
 ---
 
 # Profile imaging pixelizations at production over-sampling

diff --git a/tasks/interferometer_large_mask_aris.md b/tasks/interferometer_large_mask_aris.md
@@ -347,3 +347,16 @@ Not worth it (measured on the 125×125 setup unless stated):
 - **Decision record** [`direct-conv-symmetric-gate`](../decisions/direct-conv-symmetric-gate.md) added for the human decision of 2026-10-16 on PyAutoArray#635 (ymin ordering, gate 0.5 → 0.7).
 - **Human decision 2026-10-17 (chat):** interferometer CPU work is exhaustive and closed. The route-constant refit (RAL FFT/full `direct_conv` 0.68 measured vs 0.89 predicted) is deferred, not filed. Curvature preload (L3) parked: "we will come back to it soon".
 - **State:** all L-levers are shipped or parked. This task stays active only until #636 and #425 are merged and released; merge is the human's.
+
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [PyAutoArray#636](https://github.com/PyAutoLabs/PyAutoArray/pull/636) (pending-release), [autolens_profiling#425](https://github.com/PyAutoLabs/autolens_profiling/pull/425).
+- **Issues closed:** PyAutoArray#635.
+- **Mind record:** [`direct-conv-symmetric`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/direct-conv-symmetric.md).
+- Shadow row owed in Mind `autonomy_log` (glance tier; human to confirm merged-unchanged).
+- Remaining: release of #636.
+
+### Post-/prm addendum 2026-10-17
+
+- Glance shadow row appended to Mind `autonomy_log.md` for `direct-conv-symmetric` (PyAutoArray#636 / autolens_profiling#425): `merged-unchanged`, stage 1 (human answer 2026-10-17).
+- Human decision 2026-10-17: the remaining Aris levers (route-constant refit, FFT pruning, curvature preload) stay as research; not enough return to implement for now.
diff --git a/tasks/interferometer_nnls_memo_scattered_stream_guard.md b/tasks/interferometer_nnls_memo_scattered_stream_guard.md
@@ -44,6 +44,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Human decision 2026-10-17 (chat):** interferometer CPU work is exhaustive and closed. Status → parked.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#435](https://github.com/PyAutoLabs/autolens_profiling/pull/435).
+- **Issues closed:** autolens_profiling#426.
+- **Mind record:** [`interferometer-streaming-memo-after`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/interferometer-streaming-memo-after.md).
+
 ---
 
 # fnnls warm-start memo: stop it slowing scattered evaluation streams (interferometer CPU, autolens_profiling#332)

diff --git a/tasks/interferometer_streaming_scaling.md b/tasks/interferometer_streaming_scaling.md
@@ -54,6 +54,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Human decision 2026-10-17 (chat):** interferometer CPU work closed; GPU levers filed as research, no source changes. The library candidate (chunk `transformer.image_from` in `apply_sparse_operator`) stays unfiled. Optional measured 2e8 rows recorded as research, not a task. Status → parked.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#435](https://github.com/PyAutoLabs/autolens_profiling/pull/435).
+- **Issues closed:** autolens_profiling#368.
+- **Mind record:** [`interferometer-streaming-memo-after`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/interferometer-streaming-memo-after.md).
+
 ---
 
 # Campaign: interferometer streaming (array-free) vs in-memory — memory and time scaling to 2e8 visibilities

diff --git a/tasks/interferometer_w_tilde_fft_size_levers.md b/tasks/interferometer_w_tilde_fft_size_levers.md
@@ -35,6 +35,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Human decision 2026-10-17 (chat):** interferometer GPU levers — "file wiki research but not make source code changes for now". The PyAutoArray `apply_operator` JAX pruning prompt is NOT filed; the prototype (`apply_pruned` / `curvature_pruned` in `scripts/interferometer/delaunay/fft_size_levers.py`) stays in the project. Route-constant refit on CPU deferred (CPU lane closed). Status → parked (research filed).
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#433](https://github.com/PyAutoLabs/autolens_profiling/pull/433).
+- **Issues closed:** autolens_profiling#429.
+- **Mind record:** [`ral-route-constants-fft-levers`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/ral-route-constants-fft-levers.md).
+
 ---
 
 # Interferometer W~ curvature matrix is FFT-bound on the mask extent: pruned padded FFT and real-space pixel scale on the A100

diff --git a/tasks/point_source_image_plane_gpu_breakdown.md b/tasks/point_source_image_plane_gpu_breakdown.md
@@ -51,6 +51,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Human decision 2026-10-17 (chat):** image plane is "just research to file". Campaign and task → parked with this as the reference measurement; the phase-2 autolens_inference image-plane fit measurement is dropped. No baseline admission is recorded.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#434](https://github.com/PyAutoLabs/autolens_profiling/pull/434).
+- **Issues closed:** autolens_profiling#427.
+- **Mind record:** [`point-source-image-gpu-admission`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/point-source-image-gpu-admission.md).
+
 ---
 
 # Point-source A100 speed-up campaign: profile and optimize with the shared breakdown

diff --git a/tasks/point_source_source_plane_chi_squared_speed.md b/tasks/point_source_source_plane_chi_squared_speed.md
@@ -38,6 +38,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - **Human decision 2026-10-17 (chat):** source-plane speed-ups "not worth pursuing again, file as research". Task → complete; campaign closed as research. Blackjax stays parked. The Adam step-limit observation is recorded as a PyAutoFit search-settings note, not a task.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_inference#29](https://github.com/PyAutoLabs/autolens_inference/pull/29), [autolens_profiling#432](https://github.com/PyAutoLabs/autolens_profiling/pull/432).
+- **Issues closed:** autolens_inference#25, autolens_profiling#428.
+- **Mind record:** [`ps-source-gradient-admission`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/ps-source-gradient-admission.md).
+
 ---
 
 # Point-source source-plane chi-squared speed-up campaign — remaining candidates (phases 1–2e shipped)

diff --git a/tasks/runtime_cell_single_jit_gpu_warmup.md b/tasks/runtime_cell_single_jit_gpu_warmup.md
@@ -46,6 +46,12 @@ not refresh measurements or authorize compute. Historical task text is retained.
 - Task → complete (witness met at PR-open; merge is the human's). Remaining release-sweep cells gain the field at the next full sweep.
 - [Full dated review](../checkins/2026-10-17.md). Review time does not refresh measurements or authorize compute; merge is the human's.
 
+### Post-/prm 2026-10-17
+
+- **Merged 2026-10-17 (human /prm):** [autolens_profiling#432](https://github.com/PyAutoLabs/autolens_profiling/pull/432).
+- **Issues closed:** autolens_profiling#428.
+- **Mind record:** [`ps-source-gradient-admission`](https://github.com/PyAutoLabs/PyAutoMind/blob/main/complete/2026/10/ps-source-gradient-admission.md).
+
 ---
 
 # Runtime cells' A100 `single_jit` includes the post-compile warm-up — source-plane 0.642 ms vs a steady 0.267 ms on the same node