Caveats that I took the tests in rural town in the Midwest over a decades ago and didn’t think I saw anything unexpected for what it was trying to do: This is not first hand knowledge.

Past versions have been found to have questions that rely heavily on knowledge that wouldn’t be obtained during standard primary education. The worst I’d heard about was knots used while boating. If you’re privileged enough to have that exposure, you’re failing that one. No other reason to expect it there, so no reason to learn that.

My pixel has four options from that screen, one being lockdown, which appears to be similar.

Having toured a few dairy plants in Missouri while job hunting a few years back, the pasteurization is doing some really heavy lifting. Very concerned about hormones in milk. But not so much about open vat homogenizers in excessively damp rooms or cleaning the road grime out of the hoses on trailer trucks bringing the milk in. Give it quick 48hr incubation on TSA and call it good. “Milk expires too fast to bother with any other testing” Wild the F in FDA is apparently the least fucks given part. “Post in notes in the room where they’re milling stainless steel down for stethoscope bodies?! Those could be uncontrolled documents!”

So I can’t help with exact physics too much or exact electrical. Just a manufacturing engineer with too many hobbies.

At a basic level, you’ll want to decide what you’re making. While similar, the specifics will decide what you need exactly. A coilgun or mass driver uses electromagnetic coils in a series, 1 -> 2 -> 3 ect. Each coil is powered one after the other, with sufficient uptime and delay before the next to pull a ferrous carrier or projectile forward. Moments before it reaches the powered coil, that coil should shut down, moving the field further and continuing the acceleration. Note, as you accelerate the payload you will see a shorter uptime and delay at subsequent coils.

Of note: How you’re moving the payload is important. If you’re using a carrier, you will need to either account for its separation from the payload (consider a sabot in a shotgun shell, though many other designs likely exist) or its deceleration before the end of the line to avoid its self destructive impact. A carrier less payload needs some means of moving along with minimal friction, and must be ferrous so that the induced fields will interact with it. Most I’ve seen made used a non conductive tube polished smooth inside with each coil wound around the tube itself. Electronics on the outside, launching a steel ball bearing. Easily procured capacitors from cameras set up to each coil and confirmed to pop off in series. Either a LONG tube, or potentially hazardous electrical charges. Please consult a real electrician or engineer before using higher power inputs.

It’s certainly been done quite often, but doing anything particularly impressive will be a bit of a feat without significant input.

A rail gun is actually far easier to make, if even less impressive without massive power inputs (potentially millions of amps to achieve the theoretical velocities a railgun is associated with) and precision manufacturing. Instead of creating a coil, a neat bit of physics is abused. Functionally, your design will resemble an elongated H, with the bottom legs extending as long as you’d like. Power source wires to the tops of each leg. The center portion serves to bridge the rails and complete the circuit and will also need to move up and down them freely. What this creates is a magnetic field between the center bridge and the power inputs on top. This field will propel the center down the rails. Every bit of rail behind the connection will generate a magnetic field as it does so. No need to time coils.

Sled is a must, as the center must contact the rails the whole time. The more magnetically reactive you can get the sled/payload, the easier acceleration becomes. Adding magnets to the sled isn’t uncommon.

As with the coil gun, getting anything impressive becomes quite a feat. A demonstration of what’s happening is fairly easy.

We’ve done childrens demonstrations as part of the company outreach like this. A magnet and paperclips will suffice. I would suggest a railgun and you start there, as the demonstration is easily found online and scale up from there. Uni project, not Naval bombardment being your goal.

I cannot stress enough. Please consult with someone knowledgeable in electronics if you start moving into higher power inputs. A demo model mishap with 9V is a tingle. Running 120V gets serious very quickly.

Are you trying to build one, interested in the specifics of the electromagnetics, or just looking for a general overview?

Lots of difference between a wiki level overview, the physics involved, or the engineering behind making a theoretical item function in reality.

I don’t know enough about copilot as work has made it optional for mostly accessibility related tasks: digging through the mass of extended Microsoft files in teams, outlook, OneDrive to find and summarize topics; record meeting notes, not that they’re overly helpful compared to human taken notes due to a lack of context; and normalizing data, as every power BI report out is formatted as it’s owner saw fit.

Given it’s ability to make ridiculous errors confidently, I don’t suppose it has the memory to be used more like a toddler helper? Small, frequent tasks that are pretty hard to fuck up, once it can reliably do these through repetition and guidance on what’s a passing result, tieing more together?

If you don’t mind me asking, what do you do and kind of AI? Maybe it’s the autism but I find LLMs are bit limited and useless but other use cases aren’t quite as bad Training image recognition into AI is a legitimately great use of it and extremely helpful. Already being used for such cases. Just installed a vision system on a few of my manufacturing lines. A bottling operation detects cap presence, as well as cross threads or un-torqued caps based on how the neck vs cap bottom angle and distance looks as it passes the camera. Checking 10,000 bottles a day as they scroll past would be a mind numbing task for a human. Other line is making fresnel lenses. Operators make the lenses, and are personally checking each lens for defects and power. Using a known background and training the AI to what distortion good lenses should create when presented is showing good progress at screening just as well as my operators. In this case it’s doing what the human eye can’t; determine magnification and defraction visually.

Pushing Little Children With Their Fully Automatics!

Didn’t make this, so only a guess, but the concept of plants “communicating” via mycorrhizae is probably why it’s there.

What kind/where do you get leather from?

I’ve been using veg tan from Tandy I’ve had better luck with smooth cuts on oil tan or chrome tan, but they’re not quite suitable for boiling

Or maybe it’s too fine an edge? I’ve a retired scalpel and a plethora of blades from my old university job I’ve repurposed.

Most of my stuff has been utilitarian or functional enough it’s not mattered. An nice abrasion/heat resistant shop apron, or renfaire bags

I made a few pieces from scrap to test a book mark and they ended up rough.