True story: I have extensive experience in Resource Allocation Algorithms/Systems, so several years ago I was offered to be chief engineer for developing a novel energy distribution system 1/ cleantechnica.com/2021/09/09/tes…
2/ I started learning about the project (which had already started along with a subcontractor) and the problem at hand.
Real interesting project!
Client was a foreign utility company which supplied power to lots of varied customers - from retail households to large industry
3/ They had to balance loads against powerplants of various types - some costly and slow to spool up and down, others with minimal control over output etc.
This was a new field, real interesting. Before I gave my answer, something happened.
4/ The subcontractor was a strategic partner of the company that offered me the position.
It was decided by top management to pass them all the IP required for the project (huh? who does that)
5/ Soon after, the SUBCONTRACTOR signed the deal for the energy distribution system, leaving the company where I was offered the position completely outside 😮
Of course, the position was gone. Luckily I hadn't left my previous one like I meant to
Management decided not to sue!
6/ Years later I found out that the managers from main company retired to work with hefty salaries at the subcontractor
🤔
But I'm sure the "Transfer All IP" decision was "clean", right?
As were grossly inflated (IMO) deals the subcontractor was paid to work on other projects🤷♂️
7/ If not mistaken the subcontractor failed to deliver or got the rug pulled away later👎
Had it turned otherwise, I could have developed the first AI-driven power production/distribution system anywhere!
I have no idea how good I would have done, but would've given all I got!
8/ Been close to the topic ever since that failed job offer. It's a huge multi-factorial engineering problem with huge financial implications, which begged to be solved.
So happy to see that the best such system was made by a company I love, and it KICKS ASS.
WHY X AND S ARE STILL PRODUCED
[The hidden reasons]
With 3 and Y providing most of the automotive revenues and Cybertruck providing a bigger halo, why does Tesla even bother making them?
This thread will show the hidden reasons that nobody's talking about.
1/🧵
2/ @FutureAZA nailed it in his post, so a good place to start is there.
S and X are profitable, so Tesla's still making them. However, their gross margin is MUCH HIGHER than one would expect.
How's that, you ask?
That's because unlike all other models, S & X production line should be fully amortized. While other models have to pay for their production lines, S&X line is almost free
2b/
BTW That's one of the reasons GM continued making the Bolt, and one of the main reasons Ford and GM lose so much money on their EVs.
Bolt's production line is mostly amortized, while for Mach E, LyrIQ etc very few units have to carry the amortization of expensive new factories and lines
Shortly before Tesla announced that they're bringing Elon's Comp Plan to shareholder's vote, I sent my patrons a video containing predictions
They requested public debate, so here we are
1/🧵
Despite widespread belief that tesla was about to appeal, I suggested considering the option of them bringing it to vote (which was later announced) and of the vote failing (explained later in thread)
In this case, I expect Elon to leave, ONLY IT'S NOT THAT SIMPLE...
As some of you know (and shown with numerous examples in my Highland video), Elon is a disciple of Henry Ford. This could be important.
In my latest video I told the story of how the court forced Ford to pay shareholders dividends instead of investing in growth. Ford hated that
WELL, THIS AGED WELL...
When Tesla had its 2022 AI Day, a group of robotics experts and nerds mocked Tesla's efforts on a live-thread led by @ctwy
Now that @Tesla_Optimus is about to enter productive work in factories, I thought it's a good time to check how well their expert predictions aged
➡️Here they mocked the presentations "as a grad student/TA doing a class presentation for a bunch of undergrads. Feels very 101."
(Considering the progress we saw, I guess the grad student and undergrads learned pretty fast😎)
1/🧵
@GoingBallistic5 @ICannot_Enough
2/ ➡️They mocked Tesla's choosing 5 fingers, when two finger pinchers or vacuum cups (like what her employer @PlusOneRobotics does) are better for warehouse duties
(could be, but let's see them use vacuum cups to operate power tools)
3/ They didn't get why Tesla favored a humanoid form, when purpose-built shapes are better
(Yeah, better for one specific task it was purposely designed for)
TESLA GIGACASTING UPPER BODY + DIFFERENT US HL M3?
In this thread I'll explain (w. video) why "Entire underbody in one piece" rumor is wrong, as is "Entire car in one shot"
In fact, I see new rumor as GIGACASTING the UPPER BODY. I also see a possible DIFFERENT US HL M3
1/🧵
2/ Patent is for casting entire car, but is that what Tesla is doing now? I think not!
"Entire underbody" prevents Gen3's parallelization. It is the exact opposite of "let's let robots huddle in parallel around each body section (front, mid, rear), then attach them together"
3/ Video here explains the UPPER BODY.
This was predicted before investors day.
Ignore model name guesses and what will be on Highland and what on Gen 2, I'm still sure I got most of the tech right.
I got a SNEAK PEEK into TESLA's upcoming RIDE-HAILING APP!
You did too. You're using it right now!
@Uber+@Lyft's main value for @Tesla is millions of users on their app
But click a button here on X for a Robotaxi, and BOOM! Uber's dead! 1/ https://t.co/by6K4cLatt