AI Flow
Public Made by Adomby adom
Adom's AI Flow: a tool to help the AI follow all of the steps it takes to build a board.
main
AI Flow run: Claude Fable 5.1 on board-0
Delivered at 2026-09-14T03:03:07Z: 53.4 min of AI time from the prompt to "done, here is your video" (53.4 min on the wall clock). Since then 401.2 min more of AI time over 24 follow-up prompt(s), listed under Sessions; none of it is in the delivery number.
Prompt at 2026-09-14T02:09:45Z (UTC). Flow: board. Every number here is from the run's append-only ledger (run.jsonl). AI time is the time the AI was working on a prompt: from the human's prompt to the AI's done, with any gap over 15 min where no command ran (the human away) cut out.
The video
The final cut of this run, 120 s: only the moments something moved, every step in order, the AI's analysis drawings fullscreen, the run and step clocks in the corner, the words cut to the picture.
Side by side with Codex (Astra)
Claude Fable 5.1 on the left, Codex (Astra) on the right, step by step in flow order, 186 s: each run's own clocks in its own corner, both AIs' drawings side by side, the words cut to the picture.
Sessions
One row per human prompt: the AI's time on it, the idle it did not get charged for.
| Session | Prompt at | Done at | AI time min | Idle cut min | What |
|---|---|---|---|---|---|
| 1 | 2026-09-14T02:09:45Z | 2026-09-14T03:03:07Z | 53.4 | 0 | the first prompt |
| 2 | 2026-09-14T03:03:10Z | (open) | 7.9 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 3 | 2026-09-14T03:28:32Z | (open) | 1.8 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 4 | 2026-09-14T11:18:58Z | (open) | 19.5 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 5 | 2026-09-14T12:06:09Z | (open) | 142.0 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 6 | 2026-09-14T15:31:53Z | (open) | 14.9 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 7 | 2026-09-14T16:11:26Z | (open) | 4.8 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 8 | 2026-09-15T12:03:43Z | (open) | 71.2 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 9 | 2026-09-15T14:28:43Z | (open) | 0.0 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 10 | 2026-09-15T14:28:43Z | 2026-09-15T14:29:25Z | 0.7 | 0 | add the fields step: adom-fields on the board, walked on camera |
| 11 | 2026-09-15T14:29:25Z | (open) | 4.0 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 12 | 2026-09-15T14:33:26Z | 2026-09-15T14:33:33Z | 0.1 | 0 | add the fields step: adom-fields on the board, walked on camera (second try, the app windo |
| 13 | 2026-09-15T14:33:33Z | (open) | 6.6 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 14 | 2026-09-15T14:40:09Z | 2026-09-15T14:41:36Z | 1.4 | 0 | add the fields step: adom-fields on the board, walked on camera (third try, a fresh pup se |
| 15 | 2026-09-15T14:41:36Z | (open) | 6.1 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 16 | 2026-09-15T14:47:44Z | 2026-09-15T14:50:12Z | 2.5 | 0 | add the fields step: adom-fields on the board, walked on camera (fourth try, the window is |
| 17 | 2026-09-15T14:50:12Z | (open) | 6.7 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 18 | 2026-09-15T14:56:54Z | 2026-09-15T14:59:24Z | 2.5 | 0 | add the fields step: adom-fields on the board, walked on camera (fifth try, pup warmed fir |
| 19 | 2026-09-15T14:59:24Z | (open) | 10.6 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 20 | 2026-09-15T15:10:02Z | 2026-09-15T15:10:37Z | 0.6 | 0 | add the fields step: adom-fields on the board, walked on camera (sixth try, filmed on Adom |
| 21 | 2026-09-15T15:10:37Z | (open) | 6.5 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 22 | 2026-09-15T15:17:06Z | 2026-09-15T15:20:22Z | 3.3 | 0 | add the fields step: adom-fields on the board, walked on camera (seventh try, the app wind |
| 23 | 2026-09-15T15:20:22Z | (open) | 13.8 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 24 | 2026-09-15T16:16:25Z | (open) | 12.7 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| 25 | 2026-09-15T21:46:19Z | (open) | 61.0 | 0 | follow-up (no prompt mark; say prompt --text next time) |
| all | 454.6 |
Per step
AI time only, the idle cut out.
| Step | AI min | AI thinking min | Tool min | Turns | Visits | Returns | Longest think |
|---|---|---|---|---|---|---|---|
| intake | 6.5 | 5.5 | 0.9 | 6 | 2 | 0 | 3.9 min |
| placement | 2.6 | 1.8 | 0.8 | 12 | 2 | 1 | 1.5 min |
| routing | 28.5 | 11.9 | 16.6 | 25 | 2 | 0 | 2.6 min |
| pours | 10.1 | 3.0 | 7.1 | 20 | 2 | 1 | 1.3 min |
| current | 1.3 | 0.0 | 1.3 | 6 | 2 | 1 | 0.0 min |
| thermal | 2.2 | 0.8 | 1.4 | 11 | 4 | 2 | 0.5 min |
| finish | 4.0 | 3.8 | 0.2 | 114 | 13 | 0 | 3.8 min |
| 3d | 5.5 | 0.0 | 5.5 | 4 | 3 | 1 | 0.0 min |
| fields | 10.6 | 0.0 | 10.6 | 14 | 7 | 7 | 0.0 min |
| models | 10.8 | 0.0 | 10.8 | 3 | 1 | 0 | 0.0 min |
| nets | 1.1 | 0.0 | 1.1 | 5 | 2 | 2 | 0.0 min |
| total | 83.0 | 26.7 | 56.3 |
Returns to earlier steps
- 2026-09-14T02:20:24Z back to placement (visit 2): router: no escape path at U5.40, U5.32, U5.14 (decoupling caps packed too close to the LQFP's pins) and J1.4 (SWD header jammed in the corner); open the escape band around U5 and move J1
- 2026-09-14T02:56:45Z back to pours (visit 2): thermal: Q2 at 69 C and Q4 at 64 C against the 60 C budget; the phase A and B spreaders fill as fragments under the driver's B.Cu routing; add B.Cu copper right of the +VBAT column at the phase contacts and a bottom band for phase A
- 2026-09-14T12:57:46Z back to nets (visit 1): first net walkthrough fired KiCad's Zoom to Selection Area tool (a substring match) and parked the editor; the bridge now matches exact labels
- 2026-09-14T13:00:15Z back to nets (visit 2): the check of the 1.0.10 bridge re-fired the Zoom to Selection Area tool and parked the editor before the second walk; 1.0.11 matches exact labels
- 2026-09-14T13:35:15Z back to 3d (visit 3): the binary's first 3D tour filmed an unpainted viewer (opened in the background, canvas never drew); the tour now shows the viewer and waits for a painted canvas before recording
- 2026-09-14T15:31:54Z back to current (visit 2): the flow now asks for the AI's own current density drawing and a walkthrough of the pours the analysis judged
- 2026-09-14T15:32:26Z back to thermal (visit 3): the flow now asks for the AI's own temperature rise drawing and a walkthrough of the pours the analysis judged
- 2026-09-14T16:11:27Z back to thermal (visit 4): the heat analysis now has two drawings: the heat flow in the copper per chip and its vias, then the board's FR4 spreading
- 2026-09-15T14:28:44Z back to fields (visit 1): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T14:33:26Z back to fields (visit 2): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T14:40:10Z back to fields (visit 3): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T14:47:45Z back to fields (visit 4): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T14:56:55Z back to fields (visit 5): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T15:10:02Z back to fields (visit 6): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
- 2026-09-15T15:17:07Z back to fields (visit 7): the flow gained the fields step: adom-fields solves the current and the heat on the board and walks it on camera
Decisions along the way
- 2026-09-14T02:12:17Z clip-start:intake-2
- 2026-09-14T02:13:50Z John asked mid-run for a screenshot per step; built it into the binary (start and end of every step visit) before continuing intake
- 2026-09-14T02:16:36Z clip-start:placement-1
- 2026-09-14T02:17:59Z place pack packed wishes alphabetically (JSON objects carry no order): fixed in the binary, wishes are now an ordered array
- 2026-09-14T02:18:38Z clip-start:routing-1
- 2026-09-14T02:20:36Z clip-start:placement-2
- 2026-09-14T02:20:48Z clip-start:routing-2
- 2026-09-14T02:23:25Z intake decision: the start board has no inner planes; GND (In1) and +3V3 (In2) planes added as board-4 and as the first two pours; spec rewritten for this placement (spec-run.json)
- 2026-09-14T02:25:47Z gate: one 0.1989 mm clearance (a /5V_EN track against U2 pad 3): the router now keeps 10 um over the cell diagonal; rebuilt, routing again
- 2026-09-14T02:29:46Z the 10 um margin choked the escapes (35 unrouted: the dilation radius crossed a cell boundary); reverted; per-net clearance override added to the router instead, /5V_EN at 0.22
- 2026-09-14T02:34:10Z found it: oval pads were modelled as polygons inscribed in the arc, 1.2 um short of the copper; arcs are now circumscribed; override removed, routing again
- 2026-09-14T02:37:10Z root cause: the escape stub from U2.2 sloped toward the off-grid escape cell from the pad centre and grazed pad 3; stubs now leave along the pad's long axis
- 2026-09-14T02:49:32Z clip-start:pours-1
- 2026-09-14T02:52:40Z GND pour starved at U6.3, the INA181's Kelvin IN- pad: the pour must not touch a sense pad at all; the pours crate now puts a keepout over every Kelvin sense pad
- 2026-09-14T02:53:48Z +VBAT block and column shared an edge on B.Cu (zones_intersect, same net, same priority): merged into one B.Cu polygon
- 2026-09-14T02:56:09Z clip-start:current-1
- 2026-09-14T02:56:14Z clip-start:thermal-1
- 2026-09-14T02:56:53Z clip-start:pours-2
- 2026-09-14T02:58:29Z run.json lost page and userMachine: the background landing saved a stale copy over them (concurrent writers); fixing save to merge
- 2026-09-14T02:58:52Z clip-start:thermal-2
- 2026-09-14T02:59:13Z clip-start:finish-1
- 2026-09-14T03:30:16Z after delivery: the clips were re-cut motion-only with a real-time clock (0.1.7 action cut) and the final video re-composed from them; the clock of record is unchanged
- 2026-09-14T11:19:41Z compose: the delivered video replaced by the composed cut (title cards and action cuts); the clock of record unchanged
- 2026-09-14T11:21:28Z compose: the delivered video replaced by the composed cut (title cards and action cuts); the clock of record unchanged
- 2026-09-14T11:32:29Z recut: 10 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T11:34:25Z compose: the delivered video replaced by the composed cut (overlays, action cuts, narration); the clock of record unchanged
- 2026-09-14T12:10:48Z recut: 11 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T12:16:41Z recut: 11 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T12:42:43Z clip-start:models-1
- 2026-09-14T12:42:55Z net +12V: 124 items highlighted
- 2026-09-14T12:57:47Z clip-start:nets-1
- 2026-09-14T12:58:02Z clip-start:3d-2
- 2026-09-14T12:58:56Z clip-start:finish-2
- 2026-09-14T13:00:29Z clip-start:nets-2
- 2026-09-14T13:00:36Z net +12V: 124 items highlighted
- 2026-09-14T13:00:41Z net +5V: 97 items highlighted
- 2026-09-14T13:00:45Z net +VBAT: 283 items highlighted
- 2026-09-14T13:00:49Z net /DRV_SHA: 108 items highlighted
- 2026-09-14T13:00:54Z net /DRV_SHB: 178 items highlighted
- 2026-09-14T13:00:58Z net /DRV_SHC: 129 items highlighted
- 2026-09-14T13:01:03Z net GND_OUT: 11 items highlighted
- 2026-09-14T13:01:08Z net GND: 269 items highlighted
- 2026-09-14T13:01:28Z clip-start:finish-3
- 2026-09-14T13:07:54Z recut: 16 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T13:12:42Z compose: the delivered video replaced by the composed cut (raw clips with the ledger's overlays, narration); the clock of record unchanged
- 2026-09-14T13:18:06Z compose: the delivered video replaced by the composed cut (raw clips with the ledger's overlays, narration); the clock of record unchanged
- 2026-09-14T13:31:57Z clip guard: the finish clip ran 30 min, past its 10 min budget twice over; stopped by the guard
- 2026-09-14T13:35:16Z clip-start:3d-3
- 2026-09-14T13:36:30Z clip-start:finish-4
- 2026-09-14T13:54:09Z recut: 19 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T13:59:49Z compose: the delivered video replaced by the composed cut (raw clips with the ledger's overlays, narration); the clock of record unchanged
- 2026-09-14T13:59:49Z clip guard: the finish clip ran 23 min, past its 10 min budget twice over; stopped by the guard
- 2026-09-14T14:22:07Z recut: 20 clips re-cut (10x and action) with this binary's rules
- 2026-09-14T14:28:05Z compose: the delivered video replaced by the composed cut (raw clips with the ledger's overlays, narration); the clock of record unchanged
- 2026-09-14T15:31:55Z clip-start:current-2
- 2026-09-14T15:32:01Z artifact analysis-image for current-2: Current density on the copper, the AI's own DC conduction solve on the filled polygons of every loaded net at its spec current: the 20 A nets peak at 230 to 320 amps per square millimetre where a one millimetre track carries a phase, the pours sit far lower.
- 2026-09-14T15:32:03Z net +VBAT: 283 items highlighted
- 2026-09-14T15:32:07Z net /DRV_SHA: 108 items highlighted
- 2026-09-14T15:32:12Z net /DRV_SHB: 178 items highlighted
- 2026-09-14T15:32:16Z net /DRV_SHC: 129 items highlighted
- 2026-09-14T15:32:21Z net GND_OUT: 11 items highlighted
- 2026-09-14T15:33:00Z clip-start:thermal-3
- 2026-09-14T15:33:06Z artifact analysis-image for thermal-3: Temperature rise over the board, the AI's own steady-state heat solve: 4.1 W from the seven hot tabs plus 3 W of Joule loss in the copper, spread by every layer, lost to the air. In still air the FET tabs sit about 100 degrees above ambient, over the 60 degree budget; in moving air the peak is 57 degrees.
- 2026-09-14T15:33:08Z net +VBAT: 283 items highlighted
- 2026-09-14T15:33:13Z net /DRV_SHA: 108 items highlighted
- 2026-09-14T15:33:17Z net /DRV_SHB: 178 items highlighted
- 2026-09-14T15:33:22Z net /DRV_SHC: 129 items highlighted
- 2026-09-14T15:33:48Z clip-start:finish-5
- 2026-09-14T15:33:55Z capture-stop
- 2026-09-14T15:36:35Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-14T15:46:44Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-14T16:11:28Z clip-start:thermal-4
- 2026-09-14T16:11:32Z artifact analysis-image for thermal-4: Heat flow in the copper, the AI's own four-layer heat solve read on F.Cu and B.Cu like the current picture: each hot tab's watts spreading along its layer and going down its vias, the vias drawn. Q5 sends 35 percent of its heat down three vias; Q3 has no thermal via within 3 mm and sends none, so it is the hottest tab.
- 2026-09-14T16:11:32Z artifact analysis-image for thermal-4: Temperature rise over the board, the same four-layer solve read as temperature on the top face, the FR4 spreading: 4.1 W from the seven hot tabs plus 3 W of Joule loss. In still air the FET tabs sit 105 to 121 degrees above ambient, over the 60 degree budget; in moving air the peak is far lower.
- 2026-09-14T16:11:45Z clip-start:finish-6
- 2026-09-14T16:11:50Z capture-stop
- 2026-09-14T16:16:08Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T12:22:53Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T14:38:04Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T14:46:08Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T14:54:49Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T15:04:04Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T15:14:39Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T15:24:31Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
- 2026-09-15T15:34:04Z compose: the delivered video replaced by the composed cut (motion only, the ledger's overlays, the AI's drawings, narration cut to the picture); the clock of record unchanged
Where the board stands
- routing: 0 unrouted after null passes
- gate: 0 new DRC errors, 0 unconnected, null inherited
- landed live: 116 of 116 route entries
- pours: 4 groups, failed []
- copper kept (mm2 per layer): {"B.Cu":3218.34,"F.Cu":2811.17,"In1.Cu":4114.64,"In2.Cu":4062.9}
- analyze-current: pass
- analyze-thermal: pass
The steps on camera
One clip per step visit, sped up 10x, with the editor at the start and the end of the visit. A caption above every clip and screenshot says what it shows.
intake, visit 2
What this step does: read the board and the spec, write the spec from the schematic if it is missing, plan
Clip: intake, visit 2, the PCB editor from 2026-09-14T02:12:17Z to 2026-09-14T02:16:14Z. Motion only: 3.9 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the intake, visit 2 clip, what the camera saw.
Screenshot: the board when intake, visit 2 ended.
placement
What this step does: place the parts for routability, current and heat; the binary packs, checks courtyards and lands moves
Clip: placement, the PCB editor from 2026-09-14T02:16:36Z to 2026-09-14T02:18:27Z. Motion only: 1.8 min of real time, 1 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the placement clip, what the camera saw.
Screenshot: the board when placement began.
Screenshot: the board when placement ended.
routing
What this step does: route to 100 percent with the spec's rules, gate with KiCad's DRC, land as undo steps; the AI takes it with take route=ai
Clip: routing, the PCB editor from 2026-09-14T02:18:38Z to 2026-09-14T02:20:25Z. Motion only: 1.8 min of real time, 1 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the routing clip, what the camera saw.
Screenshot: the board when routing began.
Screenshot: the board when routing ended.
placement, visit 2 (rework)
What this step does: place the parts for routability, current and heat; the binary packs, checks courtyards and lands moves
Why the AI came back here: router: no escape path at U5.40, U5.32, U5.14 (decoupling caps packed too close to the LQFP's pins) and J1.4 (SWD header jammed in the corner); open the escape band around U5 and move J1
Clip: placement, visit 2 (rework), the PCB editor from 2026-09-14T02:20:36Z to 2026-09-14T02:20:45Z. Motion only: 0.2 min of real time, 1 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the placement, visit 2 (rework) clip, what the camera saw.
Screenshot: the board when placement, visit 2 (rework) began.
Screenshot: the board when placement, visit 2 (rework) ended.
routing, visit 2
What this step does: route to 100 percent with the spec's rules, gate with KiCad's DRC, land as undo steps; the AI takes it with take route=ai
Clip: routing, visit 2, the PCB editor from 2026-09-14T02:20:48Z to 2026-09-14T02:47:20Z. Motion only: 26.6 min of real time, 34 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the routing, visit 2 clip, what the camera saw.
Screenshot: the board when routing, visit 2 began.
Screenshot: the board when routing, visit 2 ended.
pours
What this step does: pour every net the spec names, keepouts along Kelvin taps, thermal and stitching vias, land, measure the copper kept
Clip: pours, the PCB editor from 2026-09-14T02:49:32Z to 2026-09-14T02:55:28Z. Motion only: 5.9 min of real time, 38 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the pours clip, what the camera saw.
Screenshot: the board when pours began.
Screenshot: the board when pours ended.
current
What this step does: IPC-2221 on every loaded net, pour and via capacity, the same-layer rule
Clip: current, the PCB editor from 2026-09-14T02:56:09Z to 2026-09-14T02:56:11Z. Motion only: 0.0 min of real time, 2 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the current clip, what the camera saw.
Screenshot: the board when current began.
Screenshot: the board when current ended.
thermal
What this step does: copper and vias at every hot tab against the rise budget
Clip: thermal, the PCB editor from 2026-09-14T02:56:14Z to 2026-09-14T02:56:46Z. Motion only: 0.5 min of real time, 6 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the thermal clip, what the camera saw.
Screenshot: the board when thermal began.
Screenshot: the board when thermal ended.
pours, visit 2 (rework)
What this step does: pour every net the spec names, keepouts along Kelvin taps, thermal and stitching vias, land, measure the copper kept
Why the AI came back here: thermal: Q2 at 69 C and Q4 at 64 C against the 60 C budget; the phase A and B spreaders fill as fragments under the driver's B.Cu routing; add B.Cu copper right of the +VBAT column at the phase contacts and a bottom band for phase A
Clip: pours, visit 2 (rework), the PCB editor from 2026-09-14T02:56:53Z to 2026-09-14T02:58:40Z. Motion only: 1.8 min of real time, 7 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the pours, visit 2 (rework) clip, what the camera saw.
Screenshot: the board when pours, visit 2 (rework) began.
Screenshot: the board when pours, visit 2 (rework) ended.
thermal, visit 2
What this step does: copper and vias at every hot tab against the rise budget
Clip: thermal, visit 2, the PCB editor from 2026-09-14T02:58:52Z to 2026-09-14T02:59:10Z. Motion only: 0.3 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the thermal, visit 2 clip, what the camera saw.
Screenshot: the board when thermal, visit 2 began.
Screenshot: the board when thermal, visit 2 ended.
finish
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, still recording (started 2026-09-14T02:59:13Z); it appears here when the step ends.
Screenshot: the board when finish began.
3d
What this step does:
Clip: 3d, the PCB editor from 2026-09-14T12:03:31Z to 2026-09-14T12:04:05Z. Motion only: 0.6 min of real time, 15 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the 3d clip, what the camera saw.
models
What this step does:
Clip: models, still recording (started 2026-09-14T12:42:43Z); it appears here when the step ends.
Screenshot: the board when models began.
Screenshot: the board when models ended.
nets
What this step does:
Why the AI came back here: first net walkthrough fired KiCad's Zoom to Selection Area tool (a substring match) and parked the editor; the bridge now matches exact labels
Clip: nets, the PCB editor from 2026-09-14T12:57:47Z to 2026-09-14T12:57:57Z. Motion only: 0.2 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the nets clip, what the camera saw.
Screenshot: the board when nets began.
Screenshot: the board when nets ended.
3d, visit 2
What this step does:
Clip: 3d, visit 2, the PCB editor from 2026-09-14T12:58:02Z to 2026-09-14T12:58:25Z. Motion only: 0.4 min of real time, 6 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the 3d, visit 2 clip, what the camera saw.
Screenshot: the board when 3d, visit 2 began.
3d, visit 2
What this step does:
This clip filmed nothing usable: 8 of 8 sampled frames are one flat colour in the middle of the window (an unpainted or covered canvas). It is kept as evidence and left out of the video.
Clip: 3d, visit 2, the PCB editor from 2026-09-14T12:58:26Z to 2026-09-14T12:58:50Z. Motion only: 0.4 min of real time, 6 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the 3d, visit 2 clip, what the camera saw.
Screenshot: the board when 3d, visit 2 began.
finish, visit 2
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, visit 2, the PCB editor from 2026-09-14T12:58:56Z to 2026-09-14T13:00:16Z. Motion only: 1.4 min of real time, 3 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the finish, visit 2 clip, what the camera saw.
Screenshot: the board when finish, visit 2 began.
Screenshot: the board when finish, visit 2 ended.
nets, visit 2 (rework)
What this step does:
Why the AI came back here: the check of the 1.0.10 bridge re-fired the Zoom to Selection Area tool and parked the editor before the second walk; 1.0.11 matches exact labels
Clip: nets, visit 2 (rework), the PCB editor from 2026-09-14T13:00:29Z to 2026-09-14T13:01:14Z. Motion only: 0.8 min of real time, 11 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the nets, visit 2 (rework) clip, what the camera saw.
Screenshot: the board when nets, visit 2 (rework) began.
Screenshot: the board when nets, visit 2 (rework) ended.
finish, visit 3
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, visit 3, the PCB editor from 2026-09-14T13:01:28Z to 2026-09-14T13:32:02Z. Motion only: 30.5 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the finish, visit 3 clip, what the camera saw.
Screenshot: the board when finish, visit 3 began.
3d, visit 3
What this step does:
Clip: 3d, visit 3, the PCB editor from 2026-09-14T13:35:37Z to 2026-09-14T13:36:02Z. Motion only: 0.4 min of real time, 12 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the 3d, visit 3 clip, what the camera saw.
Screenshot: the board when 3d, visit 3 began.
finish, visit 4
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, visit 4, the PCB editor from 2026-09-14T13:36:30Z to 2026-09-14T13:59:53Z. Motion only: 23.3 min of real time, 1 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the finish, visit 4 clip, what the camera saw.
Screenshot: the board when finish, visit 4 began.
current, visit 2 (rework)
What this step does: IPC-2221 on every loaded net, pour and via capacity, the same-layer rule
Why the AI came back here: the flow now asks for the AI's own current density drawing and a walkthrough of the pours the analysis judged
Clip: current, visit 2 (rework), the PCB editor from 2026-09-14T15:31:55Z to 2026-09-14T15:32:28Z. Motion only: 0.5 min of real time, 8 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the current, visit 2 (rework) clip, what the camera saw.
Screenshot: the board when current, visit 2 (rework) began.
Screenshot: the board when current, visit 2 (rework) ended.
The AI's drawing: Current density on the copper, the AI's own DC conduction solve on the filled polygons of every loaded net at its spec current: the 20 A nets peak at 230 to 320 amps per square millimetre where a one millimetre track carries a phase, the pours sit far lower.
thermal, visit 3 (rework)
What this step does: copper and vias at every hot tab against the rise budget
Why the AI came back here: the flow now asks for the AI's own temperature rise drawing and a walkthrough of the pours the analysis judged
Clip: thermal, visit 3 (rework), the PCB editor from 2026-09-14T15:33:00Z to 2026-09-14T15:33:28Z. Motion only: 0.5 min of real time, 6 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the thermal, visit 3 (rework) clip, what the camera saw.
Screenshot: the board when thermal, visit 3 (rework) began.
Screenshot: the board when thermal, visit 3 (rework) ended.
finish, visit 5
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, visit 5, the PCB editor from 2026-09-14T15:33:48Z to 2026-09-14T15:33:55Z. Motion only: 0.1 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the finish, visit 5 clip, what the camera saw.
Screenshot: the board when finish, visit 5 began.
thermal, visit 4 (rework)
What this step does: copper and vias at every hot tab against the rise budget
Why the AI came back here: the heat analysis now has two drawings: the heat flow in the copper per chip and its vias, then the board's FR4 spreading
Clip: thermal, visit 4 (rework), the PCB editor from 2026-09-14T16:11:28Z to 2026-09-14T16:11:34Z. Motion only: 0.1 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the thermal, visit 4 (rework) clip, what the camera saw.
Screenshot: the board when thermal, visit 4 (rework) began.
Screenshot: the board when thermal, visit 4 (rework) ended.
The AI's drawing: Heat flow in the copper, the AI's own four-layer heat solve read on F.Cu and B.Cu like the current picture: each hot tab's watts spreading along its layer and going down its vias, the vias drawn. Q5 sends 35 percent of its heat down three vias; Q3 has no thermal via within 3 mm and sends none, so it is the hottest tab.
The AI's drawing: Temperature rise over the board, the same four-layer solve read as temperature on the top face, the FR4 spreading: 4.1 W from the seven hot tabs plus 3 W of Joule loss. In still air the FET tabs sit 105 to 121 degrees above ambient, over the 60 degree budget; in moving air the peak is far lower.
finish, visit 6
What this step does: finish refuses until everything is true; the AI cuts the video and delivers
Clip: finish, visit 6, the PCB editor from 2026-09-14T16:11:45Z to 2026-09-14T16:11:50Z. Motion only: 0.1 min of real time, 0 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the finish, visit 6 clip, what the camera saw.
Screenshot: the board when finish, visit 6 began.
fields, visit 6
What this step does:
This clip filmed nothing usable: rejected by the AI: filmed the wiki page window, not the app: the window match was too loose. It is kept as evidence and left out of the video.
Clip: fields, visit 6, the PCB editor from 2026-09-15T15:10:19Z to 2026-09-15T15:10:23Z. Motion only: 0.1 min of real time, 3 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the fields, visit 6 clip, what the camera saw.
fields, visit 7
What this step does:
Clip: fields, visit 7, the PCB editor from 2026-09-15T15:17:20Z to 2026-09-15T15:18:38Z. Motion only: 1.0 min of real time, 42 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the fields, visit 7 clip, what the camera saw.
Delivered
What the AI said when it delivered (the video is at the top of this page):
Done. The ESC is routed to 100 percent, poured, current- and thermal-qualified, and here is the video: every step in order, the two returns included.
Plan usage, the second cost line
Percent of each plan limit used, as the provider reports it, at the start of the run and at its end.
| Provider | Limit | At start | At end | Used by this run |
|---|---|---|---|---|
| claude | 5h | 9 % | 57 % | 48.0 % |
| claude | Fable | 13 % | 76 % | 63.0 % |
| claude | week | 7 % | 43 % | 36.0 % |
| codex | week | 0 % | 0 % | 0.0 % |
The limits are per account, not per run: every AI thread signed in as this user draws on the same numbers, so a run made while other threads were working shows their usage too. Read it as an upper bound unless this run was the only thread active.
Given back
0 issue(s) or PR(s) filed by this run's AI to make the tool better for every Adom user.
Adom processes PRs within minutes during business hours: a non-stop PR intake runs all day.