The funniest part is how familiar it is. Every Elon project starts with experts explaining why it can't work.
On Forged in America, Ravin compared Figure founder Brett Adcock to Elon Musk (@elonmusk) after PayPal: "He goes into SpaceX, he goes into Tesla. People think he's insane." Then Shane Neman (@shaneneman) explained why he backed Figure (@Figure_robot) anyway. Every robotics expert he showed it to said the actuator would never work, and he took that as the signal: "When you have this kind of consensus in the field that something's impossible, that's what deep tech is."
Now the Roadster (@Tesla) is flipping off a ramp on thrusters in front of a live crowd. It isn't sustained flight, and it isn't street-legal yet, but nobody's laughing at the ramp.
Full episode: https://t.co/l3x611qvnf
@ForgedUSAPod@RavinGandhi1@DavidMKeyes
Which "impossible" idea are the experts wrong about right now?
My father died at 44. I was 13.
For years, 44 became my deadline to get everything done. Build the companies. Get married. Have kids.
My first exit brought a day of euphoria. It did not answer what I wanted the rest of my life to look like.
Now, on the drive to school, my kids and I name something from the day before that we are grateful for. That small habit changes how I look at the day ahead.
From my conversation with Al Goldstein and Ravin Gandhi on Forged in America.
#ForgedInAmerica #Entrepreneurship #Gratitude
@ForgedUSAPod@goldstein2002@RavinGandhi1
A lion can hunt field mice all day and still starve to death.
Each mouse is easy to catch and each one costs more energy to chase than it gives back. To survive, the lion has to hunt antelope. Harder, riskier, rarer, and actually worth it.
Field mice are the entire texture of a founder's day. The email that needs a reply. The meeting that could have been a message. The small customer who takes as much time as the big one. Every one of them is finishable, which is exactly what makes them so seductive.
I ran two companies this way for longer than I want to admit. Busiest person in the building, and quietly starving.
Here is the audit I now run every Sunday. It takes ten minutes.
Open next week's calendar. Next to every block, write one letter. A for antelope, meaning if this goes well it changes the trajectory of the company. M for mouse, meaning it needs doing but the company looks the same either way.
Count the A's. If there are fewer than three, the week is already lost and it has not started yet.
Then do the uncomfortable part. Pick two M's and delete them. Not delegate. Delete. See whether anyone notices.
The antelope is almost never urgent. It does not ping you. It has no deadline. It just sits there being the only thing that matters.
Ten minutes on Sunday.
Count the antelopes.
#Founders #Productivity #Focus #Entrepreneurship #Leadership #Startups #NemanVentures
Thirty years ago I did early AI work at NASA. Three weeks ago, ten thousand AI agents solved a math problem that had been open for ninety years.
Those two sentences describe the same field. They do not describe the same universe.
The problem was Navier-Stokes, one of the seven Millennium Prize problems, the equations that govern how every fluid on earth moves. The agents ran for 88 hours on an unreleased model, exchanged close to five million messages with each other, and the proof was then formally checked by machine so nobody has to take anyone's word for it.
The total cost was several million dollars in compute. A problem that stood since the 1930s fell in under four days thanks to a swarm of agents arguing with themselves at a scale no human research group could ever match.
When I was writing code at NASA, the ceiling on what a machine could figure out was the ceiling on what a person could specify. That ceiling is gone.
I do not think most people, including most investors, have updated on what that means yet.
The problems that fall next might not be chosen by us.
#AI #ArtificialIntelligence #Mathematics #AGI #DeepTech #VentureCapital #Investing #NemanVentures
Had a good discussion on the Family Office panel at @thegeforum on Wednesday with Leonard Tannenbaum, Nithin Johnson, and Geoff Marcus.
The main takeaway was how much investment timelines dictate behavior and family office principals can more easily play in asset classes largely ignored or overlooked by fund managers.
Managing permanent capital means family offices can hold a position for 10-15 years, or simply sit in cash while everyone else chases the same three trends at inflated valuations.
Patience functions as an asset class in itself.
That freedom cuts both ways, though. Without LP schedules forcing your hand, you have to build your own rigor so that flexibility doesn't turn into passivity.
Appreciated the sharp questions from Geoff, the candid talk with Len and Nithin, and the GEF team for hosting a great event.
Four years.
That's how long the oldest @paradromics implants have been recording in sheep with the signal holding steady. No migration, spikes still clean.
This won't trend because it's a sheep study and not a patient on the front page of the Times. It might be the more important news.
BCI has lived with one tradeoff forever. Single neuron resolution or durability, pick one. Matt Angle has been saying for years that was a design problem, not physics. Four years of clean data says he was right.
Nobody should need a second brain surgery because version one wore out.
Neman Ventures is an investor and I couldn't be happier about where this team is going. Congrats @Paradromics.
Disclosure: Neman Ventures is an investor in Paradromics. Companies referenced above are representative of the firm's investment strategy and are not a complete list of investments made by Neman Ventures. It should not be assumed that an investment in the companies listed was or will be profitable. A complete list of investments is available upon request. Past performance is not indicative of future results. Investments in private companies involve substantial risk, including the potential loss of principal. The Connexus BCI is an investigational device limited by U.S. law to investigational use.
Performance or longevity. For years, it seemed like the field of BCI would have to choose between the two.
Our accumulating preclinical data suggests otherwise: four years of stable neural recording and decoding in large animal models, with steady spike SNR and no electrode array migration.
Read the full announcement here: https://t.co/PA9PVSVKEH
The people I know who seem genuinely at peace are not the ones with the most. They are the ones who decided, on purpose and usually after some kind of loss, what enough actually looks like for them. A specific number. A specific kind of week. A specific set of people at the table.
Then they stopped negotiating with themselves about it.
I am not there yet. But I have started writing it down, because anything I do not define, the market will happily define for me.
More is not a plan.
Enough is.
#Entrepreneurship #FounderMindset #Wealth #PersonalGrowth #Exits #Purpose #NemanVentures
Just released: My October 2026 Roundup of what I am Reading, Watching, Listening To including:
How Obsession Built Christopher Nolan
Physics has Misunderstood Time
AI Has Solved One of Math's $1 Million Millennium Prize Problems
More ...
https://t.co/zKR2Fu2WP6