product engineer | Hardware, software, computing
Joined September 2013
- Tweets4.3K
- Following205
- Followers209
- Likes34.2K
Never thought I’d get a reply from Peter Beck himself!!!! Jheeeezzz
This will be a huge W for rocket lab.
Same in Canada pls. Where’s my hard hat
About 60% of Americans own their home. In the state of California that number is 55% in NYC it’s 33%
IN CHINA ITS FUCKING 90% NINETY PERCENT!!! IN ALL OF FUCKING CHINA. SHANGHAI, BEIJING, GUANGZHOU, SHENZEN, CHONQKING, TAIPEI, ITS ALL 90 FUCKING PERCENT !!!
We need ghost cities. We need to turn Long Island City into Shibuya, Greenpoint into Dhaka, Staten Island into Singapore. We need cement, dry wall, sewers, I need a hard hat. I’ll build it myself.
Millpreet Kamboj retweeted
What you are actually doing here is to bribe nokia to put these jobs into Canada by paying hundreds of thousands of dollars per job from taxpayer money. What this does is to lower the cost basis of nokia per employee. This has been going on for decades, called FDI which all civil servants think is a good thing. I spent a lot of time explaining to civil servants in ottawa that its not good for our economy that American and Oversees branch offices can employ Canadians at half the cost to all the canadian companies around them due to these subsidies. We should not do them at all, they are toxic, at least in the tech sector.
It's never meat to be this way, but the situation that very often arises is: It's strictly worse inside of Canada to be a Canadian company compared to a company headquartered everywhere else.
This is a bad situation, because the fruits of the subsidized labor will accrue to the wealth of other countries and not Canada. It's tax payer money invested into locking up scarce high tech talent in jobs where they no longer contribute to the Canadian economy directly. Why
Canada is leading the global tech race.
Today's milestone strengthens our digital infrastructure, drives innovation, and delivers results for Canadians across the country.
ctvnews.ca/ottawa/article/no…
It’s about time, let’s goooo
Announcing: New @Steam Hardware, coming in 2026:
Steam Controller
Steam Machine
Steam Frame
Watch our jazzy announcement video and wishlist now: steampowered.com/hardware
Millpreet Kamboj retweeted
Announcing: New @Steam Hardware, coming in 2026:
Steam Controller
Steam Machine
Steam Frame
Watch our jazzy announcement video and wishlist now: steampowered.com/hardware
Ansys is cheating on you with an electrical engineer.
autocad it’s a cranky grandmother
solidworks will build tension with and then tell you that she has to leave and then proceeds to crash your car
freecad is a pick me girl
Fusion is a girl who teaches you bad habits
ansys loves you but won’t tell you when something’s wrong
nx won’t shut up about her sister catia
It’s the opposite.
VW is going to bed with Rivian, the cards are flipping.
It’s just starting. A new game has just begun.
See what’s happening with Bugatti Rimac?
Millpreet Kamboj retweeted
Making smart glasses frames was way harder than I thought. Here's what I learned:
Materials matter. You can't use normal plastic, because it has to be light and safe to touch your face
So we chose TR90 plastic that's made for eyewear
Fake gray-market TR90 plastic is common in China. You gotta go to the source and make sure your TR90 is legit
Glue is so important. If you put too much glue it will give issues, and add weight. Too little glue, won't stick. That's why it's better for a robot to do it (VIDEO)
If the glue doesn't match the material perfectly, then it won't stick. So we tried like 4 glues to get the best one for our smart glasses
You design the frame in 3D and then make a tool. A tool is just a metal mold that is shaped like the glasses. You put a bunch of melted plastic (TR90) in the tool, let it cool, and out pops the glasses frame. Do that 10,000 times
Tooling leaves seams in the frame, so we figured out where to put the seam so it doesn't cut into your ear when you wear it
Obviously this all must be super lightweight because the electronics are heavy
The camera needs a certain perspective, so the glasses frame angles the camera down to capture hands in videos
Throughout the design process, you can't sacrifice style
And now our Mentra Live frames are born
What could really help in production to reduce warranty claim hardware and customer service costs:
1.
The most simple non technical tool - checklists at various stations.
2.
WITHOUT OVER ENGINEERING.
Having robust mechanical quality control protocols to identify issues between steps, and post final assembly.
PFMEA(process failure mode effects analysis- aka, a risk factor points system for processes). Helps Assess where first priority on quality control should be.
In short - determine which issues are the biggest foreseen risk factors based on severity, occurrence, and detectability.
Hypothetical Example - warped plastic parts is a known issue:
The severity - would be quite high for glasses.
The occurrence - could be high if hypothetically a warp prone plastic was used.
Detectability - could be designed into the part to be high with “hidden” visual inspection features or warp sensitive aligning features that don’t work if the part warps.
If the individual part cost of this warp prone part is low, even with a low manufacturing success rate(yield), the final risk to cost of this product is not be of large significance if the defective parts never makes it into an assembled unit.
Something that can’t be designed to be “auto” detectable/quantifiable like surface defects and durability may be bigger priorities than pre final assembly “auto” detectable things and issues that can be designed or programmed out.
3.
Having machine vision based visual defect testing where possible, and human conducted tests that have been validated/passed for trainability and consistency between potential operators of the process.
4.
PCBA test benches with mirrored setups
and OTA updates available could be super useful to reduce post assembly/shipment issues. FPGAs could be useful
Making smart glasses frames was way harder than I thought. Here's what I learned:
Materials matter. You can't use normal plastic, because it has to be light and safe to touch your face
So we chose TR90 plastic that's made for eyewear
Fake gray-market TR90 plastic is common in China. You gotta go to the source and make sure your TR90 is legit
Glue is so important. If you put too much glue it will give issues, and add weight. Too little glue, won't stick. That's why it's better for a robot to do it (VIDEO)
If the glue doesn't match the material perfectly, then it won't stick. So we tried like 4 glues to get the best one for our smart glasses
You design the frame in 3D and then make a tool. A tool is just a metal mold that is shaped like the glasses. You put a bunch of melted plastic (TR90) in the tool, let it cool, and out pops the glasses frame. Do that 10,000 times
Tooling leaves seams in the frame, so we figured out where to put the seam so it doesn't cut into your ear when you wear it
Obviously this all must be super lightweight because the electronics are heavy
The camera needs a certain perspective, so the glasses frame angles the camera down to capture hands in videos
Throughout the design process, you can't sacrifice style
And now our Mentra Live frames are born
Millpreet Kamboj retweeted
here's the most basic supply chain to even start a robotics renaissance in america:
upstream materials:
rare earth separation + magnet manufacturing (sintering → machining → magnetization) (for high-torque motors)
silicon carbide boule growth (for efficient power electronics)
bearing-grade steel processing (for races & ball bearings)
copper refining & wire drawing (for motor windings & pcbs)
engineering polymer compounding (for high-strength gears & frames)
pcb fabrication (the bare boards for all electronics)
battery cell production (anode/cathode refining → cell assembly)
forming & finishing:
forging (for high-strength structural parts, joints, and shafts)
investment casting & hp die casting (for complex parts, frames, housings)
laser cutting & press brake forming (for sheet metal structures)
progressive stamping (for motor laminations)
precision cnc machining (for shafts, couplings, final tolerances)
injection, insert & overmolding (for precision polymer and structural parts)
resin transfer/layup molding (composites)
heat treat, plating, anodizing, e-coating (for strength & corrosion resistance)
components & modules:
reducers (harmonic for precision, rv/cycloidal for power)
motors/servos (winding → magnetization → encoders)
power electronics modules (sic mosfets/igbts)
precision bearings (for all smooth motion)
sealed harnesses & connectors (the nervous system)
pcb assembly (smt for board-level electronics)
system & scale:
a domestic, automated alternative to foxconn/flex
robust UL-style testing & qualification facilities
materials recycling & reclamation (rare earths, metals, polymers)
china can ship a ~$6k humanoid because these layers are co-located and run 24/7. the US has pockets of excellence, but without a dense, end-to-end industrial base, we'll be dependent on others and celebrating AI slop while they're pouring castings at 3am. not to mention our complete dependence on foreign machine tools.
America is celebrating its AI software lead. But we're missing the bigger picture.
China has more robots in production than the rest of the world combined. Unitree is selling advanced humanoids for $5,900. China has entire "dark factories" factories that run with zero humans.
Jensen Huang says the “ChatGPT moment for general robotics” is coming. China is ready. Right now, the United States is not.
New WSJ piece with @AnneNeuberger on why the US needs to get serious about robotics, fast
