Jack Wilson

making things in syd

[Jack Wilson @jacks_bench]

It did in fact turn on

[Jack Wilson @jacks_bench]

Time to see if it will turn on

[Jack Wilson @jacks_bench]

@kristapsstikuts Thanks! Me and Opus 5.5 with @BackplaneWorks

[Jack Wilson @jacks_bench]

Time to see if it will turn on

[Jack Wilson @jacks_bench]

It's a panel of nodes for a distributed CAN hardware range I'm developing. Clockwise from top left:

• Projectile speed sensor (8 IR emitter/receiver pairs in a line)
• VL53L8 zoned ToF + color sensor
• Solid-state smart contactor for LiPo batteries
• Magnetic encoder
• Servo PWM + power injector

Testing the contactor and servo board on underwater ROV I'm building

[Jack Wilson @jacks_bench]

For those that don’t want to touch the GPIO header of their Jetson

Made with @BackplaneWorks

STM32G474
3x CAN-FD
3x LSM6DSV16X IMU
+ legacy IO

[Jack Wilson @jacks_bench]

My $1,065 PCBA run got hit with $229.21 of taxes and fees at the border: 21.5%

Even after reclaiming GST, that's 11.5% in fees on goods with a 0% duty rate.

Why? I was $65 over the import threshold. It was $1,000 in 2005. It's $1,000 today.

I was born in 2005.

[Jack Wilson @jacks_bench]

Feeling like a factory worker. Also the factory owner?

[Jack Wilson @jacks_bench]

JLC X-ray inspection is pretty cool

The LGAs meant to be inspected are fully shadowed by the LQFP on the opposite side but I guess that’s ok?

[Shane to the B @ShaneBOly]

@levie explained my agent setup at dinner and my family looked at me like I joined a cult

[Jack Wilson @jacks_bench]

How the test bench started, how it’s going now.

The first time Claude flashed my STM32 I was in awe
Now it tests a bus of my smart sensors while I sleep

Sometimes it needs me to do a power cycle
I’m just the meat that presses the button

Time to get a programmable power supply

[Jack Wilson @jacks_bench]

Testing some ROV thrusters from Taobao.

1/8th the price of a Blue Robotics T500. Corrosion resistance and reliability are yet to be seen, but they definitely push water.

Follow along as I build an autonomous underwater robot with them.

[Jack Wilson @jacks_bench]

4. Thoughts

After scale up, reached 0.94 on real dataset

Main takeaway for me - how fast the feedback loop was. I’d have a hypothesis, an agent would generate new data and train - I would have a result in 45 minutes.

Datasets at https://t.co/vbTnJMtFOt

[Jack Wilson @jacks_bench]

Can an underwater robot learn to see from renders alone?

Testing on real images: First model, 0.22 AP50. A week and 100k more training images, 0.94.

This is what made the transfer stick

[Jack Wilson @jacks_bench]

3. Just let your agent rent 15 4090’s

- Decided to scale up. Initial dataset was 10k images
- With formula paying off, tried scaling up dataset size
- Rendered in IsaacSim, agents make scene layouts and rent runpods
- 106,000 images rendered on 15 GPUs in 1 hour, cost $30