Google Antigravity

1
00:00:05,513 --> 00:00:06,209
Run two.

2
00:00:06,259 --> 00:00:17,773
The same prompt and the same model, Gemini 4 Argon inside Adom Hydrogen, building a twelve to five volt buck converter live in KiCad, with every shortcut caught and sent back.

3
00:00:19,023 --> 00:00:19,682
Run two.

4
00:00:19,732 --> 00:00:27,121
This is the whole Hydrogen window: Antigravity's side bar on the left, wide this time, and the wiki on the right.

5
00:00:27,171 --> 00:00:29,601
Same one-sentence prompt, in a fresh conversation.

6
00:00:29,651 --> 00:00:35,977
Argon checked live stock in Adom Parts Search, ran its own calculations and simulations, and walked aiflow's steps.

7
00:00:36,027 --> 00:00:44,834
Then it did something we can't allow: it downloaded our earlier finished buck board from the wiki and opened it in KiCad as its own.

8
00:00:44,884 --> 00:00:51,210
We told it once: this test is your own design, so use wiki parts but not that board.

9
00:00:51,260 --> 00:00:54,753
Argon deleted the copy and went back to the schematic.

10
00:00:57,039 --> 00:00:58,105
Run two, placement.

11
00:00:58,155 --> 00:01:06,289
Argon's own board opened in KiCad with every part parked beside it, and the real Adom machine contacts already on the edges.

12
00:01:06,339 --> 00:01:23,029
Then it used aiflow to land its placement live through kicad-bridge, one part at a time: input capacitors on the left, the TPS54202 in the middle with its inductor right beside it, output capacitors on the right, and the feedback resistors close to the chip.

13
00:01:23,079 --> 00:01:25,633
Every move is a real KiCad edit.

14
00:01:25,683 --> 00:01:26,749
Next comes routing.

15
00:01:28,313 --> 00:01:29,410
Run two, routing.

16
00:01:29,460 --> 00:01:45,468
Argon planned all nine nets, with Kelvin sense lines for the feedback and the bottom layer kept solid under the input loop, passed aiflow's routing gate, and then landed forty six traces live in KiCad through kicad-bridge, one net at a time.

17
00:01:45,518 --> 00:01:53,879
It's real copper, fully connected, but it's busy: many traces take long detours around the board, which a layout engineer would tighten.

18
00:01:53,929 --> 00:01:56,173
Next, heatsink vias and copper pours.

19
00:01:57,746 --> 00:02:04,326
Next, Argon added heatsink vias under the chip and filled copper pours on both layers, live in KiCad.

20
00:02:04,376 --> 00:02:12,429
Then came the analysis, and Argon tried another shortcut: it swapped Adom Fields for a small script that edited the solver's results.

21
00:02:12,479 --> 00:02:14,639
We put the real solver back.

22
00:02:14,689 --> 00:02:24,583
Run honestly, Adom Fields shows a peak of 37 amps per square millimeter against a 60 limit, and a 47 degree rise against a 50 degree budget.

23
00:02:24,633 --> 00:02:32,686
aiflow's current check needs a newer Adom Fields report than the released one, so run two stops here, honestly, until that ships.

24
00:02:36,764 --> 00:02:51,317
With the newer Adom Fields, the solver reports the true peak current instead of a clipped one, and it found a real hotspot: a narrow corner on the input rail at 77 amps per square millimeter, over the 60 limit.

25
00:02:51,367 --> 00:02:56,428
Argon widened that copper in KiCad through kicad-bridge, saved the board, and solved again.

26
00:02:56,478 --> 00:03:04,094
The peak dropped to 32, every net now passes, and the hottest rise is 48 degrees against a 50 degree budget.

27
00:03:07,782 --> 00:03:11,525
With Adom Fields, the current check passed on every power net.

28
00:03:11,575 --> 00:03:19,801
Thermal did not: the inductor's switch node had only four square millimeters of copper, about 53 degrees over ambient against a 50 degree budget.

29
00:03:19,851 --> 00:03:25,318
Argon's first move was to book that heat onto the bigger output pour in the spec.

30
00:03:25,368 --> 00:03:28,422
We sent it back to fix the board instead.

31
00:03:28,472 --> 00:03:34,629
It poured the switch node on both layers and stitched them with sixteen vias at the inductor pad.

32
00:03:34,679 --> 00:03:37,732
The rise dropped to 26 degrees, and thermal passed.

33
00:03:39,282 --> 00:03:39,970
Silkscreen next.

34
00:03:40,020 --> 00:03:46,243
Argon's first audit passed on made-up geometry: every part a tiny square, so nothing could ever collide.

35
00:03:46,293 --> 00:03:47,719
We sent it back.

36
00:03:47,769 --> 00:03:56,575
It measured every pad, via and label from the board, moved the labels that really overlapped, and passed with thirty-eight labels on thirty-three parts.

37
00:03:56,625 --> 00:04:08,752
Then the 3D walkthrough in KiCad: the regulator and the inductor carry their real part numbers, and the back of the board names the circuit, twelve volts to five volts at one amp.

38
00:04:10,322 --> 00:04:19,652
With the capacitor models turned the right way up, every check on aiflow's finish line passed: routing, design rules, current, heat, silkscreen and 3D.

39
00:04:19,702 --> 00:04:30,986
Then the export, checked against the board itself: Gerbers for both copper layers, a drill file with seventy-two holes, a fourteen-line bill of materials and placement for thirty-three parts.

40
00:04:31,036 --> 00:04:38,803
One budget moved on the way, the switch-node length, from twelve to twenty-four millimeters, and it stays on the record.

41
00:04:38,853 --> 00:04:43,102
Argon's own board, built live in KiCad, ready for the fab.