Brilliant dad, husband, chip designer, teacher, and scientist. Healthy kindness is a top goal: achieved by fasting, cold exposure and intermittent hypoxia.
@insane_analyst I will avenge this. #tufteproud! I recall well removing pitch (nm) off of a technical pub when I was young. Still feel a lil shame that i could not pushed back then. Real science is so much better. I wish they showed the bist, fmax vs tdp with units, but this is crap.
@lithos_graphein Software is hard. I think I heard you say this whole ai mega investment is kinda trash too. Is that right? Is it a boom or a bubble? I made some stuff i like so far but i am not yet feeling the singularity.
@jwt0625 When I was asked to remove physical units from graphics for publications has been a sad day and it has happened to me multiple times. Is it better than no publication at all?
@SirMaximusIX FEL etendue is not well matched to the illuminator entrance pupil which forces more efficiency losses that are often not fully counted when considering such a system. polarization is more valuable at higher NA. Interesting times.
@lithos_graphein That is such a good skit. I love that nobody knows how many yard in a mile… they have their own immersion now right? They don’t need ours 😭
@matthewjgunton Redundancy is grand of course. No way to repair a power to ground short even on a 215mm die. What is the wavelength graph in the last figure? Adding redundant circuits costs power so clean fabs are still a thing.
A lot of people quote tweeted this as 1 year anniversary of vibe coding. Some retrospective -
I've had a Twitter account for 17 years now (omg) and I still can't predict my tweet engagement basically at all. This was a shower of thoughts throwaway tweet that I just fired off without thinking but somehow it minted a fitting name at the right moment for something that a lot of people were feeling at the same time, so here we are: vibe coding is now mentioned on my Wikipedia as a major memetic "contribution" and even its article is longer. lol
The one thing I'd add is that at the time, LLM capability was low enough that you'd mostly use vibe coding for fun throwaway projects, demos and explorations. It was good fun and it almost worked. Today (1 year later), programming via LLM agents is increasingly becoming a default workflow for professionals, except with more oversight and scrutiny. The goal is to claim the leverage from the use of agents but without any compromise on the quality of the software. Many people have tried to come up with a better name for this to differentiate it from vibe coding, personally my current favorite "agentic engineering":
- "agentic" because the new default is that you are not writing the code directly 99% of the time, you are orchestrating agents who do and acting as oversight.
- "engineering" to emphasize that there is an art & science and expertise to it. It's something you can learn and become better at, with its own depth of a different kind.
In 2026, we're likely to see continued improvements on both the model layer and the new agent layer. I feel excited about the product of the two and another year of progress.
Do not miss this great opportunity for budding scientists! Learn how to use a Scanning Electron Microscope at the Exploratorium Museum in San Francisco! Witness an amazing microscopic world with your own eyes.
This special event is scheduled for the following dates/times and offered free with museum admission:
Wed, Jan 28 2026 • 11am - 12pm
Sun, Feb 1 2026 • 11am - 12pm
Wed, Feb 4 2026 • 11am - 12pm
Sun, Feb 8 2026 • 11am - 12pm
Wed, Feb 11 2026 • 11am - 12pm
For more details, visit the Exploratorium web site:
https://t.co/voMjRGlvc1
@lithos_graphein Having a flat sample is definitely good for high na euv litho. See my upcoming paper at spie al to see how to leverage design to help make the die and scribe flatter while making overlay accurate.
@jonmasters @bexcran @IDisposable Rolling coal is also a thing. I am all for a Dyson sphere but we ain’t there yet. So i would be inclined to bet against data centers in space until well after we have sub solar mass fusion… Let’s keep going on the ai for route/reticles/stco/ drc though amiright?
@ExplorationOnK7@IanCutress The metal pitch for “28nm” was 90nm so I am not sure the lie was absent there. The challenge to node naming started in the gate length over scaling era of 1998-2008. But now days it is tricky to follow. Many scaling boosters are one time benefit so we need to keep inventing.