Casey Handmer is the founder and CEO of Terraform Industries, a company focused on making synthetic fuels from solar energy, with a long-term vision of terraforming Mars. He’s a former NASA researcher, prolific blogger, and someone who has spent years thinking deeply about industrialization, energy, and how civilizations scale. This conversation ranges from the physics of heating Mars to why Rome never industrialized, from the psychology of founders to the demographic collapse facing China and the West, and from the future of AI to the mundane horror of leaded aviation gasoline still in use today.
Terraforming Mars
Mars is the most Earth-like planet — right gravity, right day length, reachable — but it’s cold. The main problem to solve is heat.
The most practical approach is mass-producing solar cells on Earth and launching them to Mars orbit to reflect sunlight onto the surface, essentially barbecuing the planet. Repurposing an iPhone production line at Foxconn could warm Mars by multiple degrees in about 10 years.
An alternative is setting up an autonomous factory on Mars that produces 35 kg/second of 10-micron nano-antennas that act as an ultra-powerful greenhouse gas, dramatically increasing thermal retention.
Nuking the poles is impractical — it would require thousands of gigaton-scale nukes per year, which is vastly more expensive than solar.
Why solar will be our main power source long-term
The world produces about 10 terawatts of electricity today. Elon Musk’s argument is that 173,000 terawatts of solar power hit Earth constantly, and not capturing it is an offense. No other energy source can scale to that level.
Per capita energy consumption has increased ~100x since the industrial revolution and could easily increase another 100x.
Solar panel production capacity already exceeds 1 terawatt annually. The transition is already happening without needing a dictator — capitalism is doing it. Environmental regulations may slow deployment by a year or two, but the outcome is not in doubt.
How to create more Elons
The PayPal mafia spawned many startups, but mostly in software. Elon was unusual in directing his energy and capital toward hardware — rockets and electric cars — in 2002-2003, when that required enormous sums and was considered almost certainly doomed.
Industrial titans like Henry Kaiser didn’t get started until their late 50s and relied on tight-knit groups of equally ambitious collaborators (his “Six Companies” built the Hoover Dam and many other projects). The lesson: get five or six extraordinary people around a table trying to impress each other for 10 years, and they can achieve almost anything.
There’s a similar cluster forming now in Los Angeles and El Segundo — young, ambitious builders who celebrate choosing hard paths over safe ones.
Many SpaceX alumni who leave don’t start companies because they’ve already proven themselves and lack a new “itch” — a thing that must exist that only they can build. Motivation follows action; you don’t wait to feel motivated, you start doing the thing.
Starting companies that won’t be built if you don’t build it
Many ex-SpaceX engineers succumb to the “siren call of venture capital” — writing checks to other founders instead of building themselves. This is a different and less critical skill set.
The bottleneck is people willing to eat glass for 10 years to build something ambitious that the future must contain. The US has experts in reindustrialization who should each be building 100 fully automated factories.
The pattern of improvement: your first attempt is bad, your tenth is much better, your thirtieth reveals how much you still suck — but your trajectory is steep.
How Casey found his mission
Around 2016-2017, Casey was working on supersonic electric flight and writing books on Mars industrialization. He realized that synthetic hydrocarbon production could ride the wave of cheap solar on Earth.
The perennial problem with space companies is that no one in space is spending money on anything. But on Earth, people spend $10 trillion a year on oil and gas — a torrent of value you can reach out and grab.
Long-term, he wants to terraform Mars and make synthetic hydrocarbons there. But the boot loader for giant solar installations is making synthetic fuel on Earth.
Playing Factorio IRL
Casey doesn’t play computer games. His attitude: if you have the financial resources, just buy a factory. Tens of thousands of factories in the US are for sale or functionally defunct. Just go buy one and build something.
Genealogy and personal history
Casey’s surname is Handmer, very unusual. He’s done extensive genealogical research and found peasant stock through and through — no famous ancestors. Most ancestors before the 1800s left no trace except genetic analysis placing them within 50 miles and 20 years.
His first generation in Australia: 9 of 11 children died in childhood. The next generation: 10 of 11 survived. Life got better.
He spent time in the Russian Far East as a teenager and early 20s — hiking in -20°C, climbing erupting volcanoes, hitchhiking, sleeping rough. He was drawn to the anarchic, neglected region far from Moscow where indigenous people and survivors live with extraordinary grit.
What role suffering plays in greatness
Suffering alone doesn’t produce greatness — if it did, we would have industrialized millions of years ago. But many great founders had enough adversity to develop a chip on their shoulder without it destroying their productive capacity.
Casey actively seeks out adversity — he organized a “suffer fest” in the mountains for hardware founders, where participants sleep on the ground, get cold, and climb rocks. The point is to show his children that life gets better or worse depending on your perspective.
How to succeed at MrBeast
Casey has watched one or two MrBeast videos and greatly admires the art form. More importantly, he read MrBeast’s onboarding memo and routinely recommends it.
The memo’s core insight: in bad companies, both employer and employee assume their responsibilities end at hiring. The employee is owed their salary regardless of outcome; the employer expects a forever minimum of work.
In a great company, employees must keep delivering and enjoy delivering to progress. Employers should want their people to level up so much that they could leave and start their own company — that’s how you know you’ve succeeded.
This is what PayPal did well: everyone saw themselves as the main character.
Studying the great builders
Casey has been systematically reading first-person memoirs and biographies of industrial titans to understand how they actually operated:
Leslie Groves (Manhattan Project): Now It Can Be Told — paragraph after paragraph of catastrophic problem, solution, catastrophic problem, solution.
Henry Kaiser: biography by one of his chief lieutenants.
Hyman Rickover: two books, one by his number two, one posthumous autobiography.
He wants the same quality of insight into Elon — how the sausage is actually made. Millions of articles exist, but almost none answer “how does he actually do it?” Elon is likely to live into his 90s, and when he stops being productive, people will wonder how they captured lightning in a bottle. Someone should publish his internal memos (with permission) the way Bezos’s shareholder letters were published.
Why Tesla succeeded where others failed
Tesla is the only electric car company to achieve sustained positive free cash flow — total revenue earned exceeds total money raised from investors. Every other EV startup (Rivian, Lucid, and many that never shipped a car) is still losing money on the same trajectory.
In space launch, at least a dozen well-funded companies founded by senior SpaceX alumni have failed. The only other success is Rocket Lab, which had a separate genesis (Peter Beck didn’t come from SpaceX).
Example of Peter Beck’s brilliance: when SpaceX undercut Rocket Lab with Transporter rideshare missions, Rocket Lab responded by buying satellite bus companies and now builds the spacecraft for 50-66% of satellites on Transporter missions — making more per launch than SpaceX does.
How Elon makes things happen
Elon is singularly good at finding the right KPI to optimize and communicating it clearly to employees. This came from long practice — early SpaceX was rough, and he learned on the job.
His learning method: when he started SpaceX, he bought all rocket textbooks in English, then taught himself German and Russian to read those textbooks too. Then he could walk up to experts like Tom Mueller and say “page 462 of Rocket Propulsion Elements, explain this.”
Casey believes Elon would not be out of place among the industrial titans of 100 years ago, just working on more advanced technology.
Elon has accepted his own suffering and found a productive way to deal with it. Casey wouldn’t wish that level of misery on anyone, but given that he is that way, he’s made all our lives better.
Grock vs. other AI labs
Casey switched from Claude to Grock as his daily driver LLM. The quality is similar, but he objects to Claude’s “hall monitor” post-training that lectures users about what they’re allowed to ask (e.g., questions about uranium-235).
Grock’s advantage is shipping a barely defensible product and iterating quickly rather than spending years trying to fix corner cases in advance.
All LLMs are mediocre at physics. Casey would like to help them be better.
Nuclear reactors on Mars — a physics puzzle
The major constraint for nuclear reactors in space or on Mars is heat disposal. On Earth, you have a giant lake as a cold sink. In space, you need radiators, which work poorly at low temperatures and must be enormous.
With radiator-limited design on a Starship-sized reactor, you get maybe 5-10 MW — not much when per capita power needs in space are ~100 kW.
Using Mars’s thin, cold atmosphere as the cold source with a giant near-supersonic fan (9m diameter, essentially a jet engine) increases output to ~300 MW. The exhaust is semi-radioactive but Mars is already radioactive.
Alternatively, put the reactor inside the habitat — it keeps everyone warm and the habitat keeps the reactor cold. The US Navy does this with hundreds of reactors inside submarine hulls.
A 100,000-person colony would need ~500 square miles of habitat area to dissipate heat — Montana ranch style, with farms, factories, forests, and ski slopes under the dome.
Building a Mars base
Starship can do single-stage return from Mars, needing ~5.5-6.5 km/s total delta-v. But since Starships are now stainless steel and cheap, most will go one-way and be scrapped or repurposed on Mars.
After 15 years of scaling up, you might launch 100,000 people per window. If everyone from 15 years prior wants to return, that’s only ~100 people — trivial.
NASA’s Mars Sample Return would cost ~$150 billion per kilogram. Scaling that to a person (500 kg with consumables) = $55 trillion. Obviously needs to be much cheaper.
If launch cost is ~$1,000/kg, flying home requires ~$500K to Mars and ~$50M back (if you have to bring all propellant). But you can make propellant on Mars with enough energy — gigawatts of it.
Solar is probably the key: cover arbitrarily large areas of Mars with solar arrays until marginal power cost is no longer the constraint.
O’Neill Cylinders vs. Mars
Bezos favors O’Neill cylinders (artificial habitats in space) over Mars colonization. Casey’s view: O’Neill cylinder proponents underestimate the convenience of having infinite rocks and raw materials right outside.
Mars landing sites have solar power, ice, iron-rich rock, and possibly other minerals. If not, you can grind up basaltic rock and separate it atom by atom — harder than mining concentrated ore, but doable.
Government strangling innovation
The set of people with the skills to perform desperately needed government reforms does not intersect with the set of people who would ever want to do it. Those skills pay far more in the private sector and yield more political influence.
Casey’s experience working with the government: they help at their own expense with no expectation of reciprocity. The problems are genuinely very hard.
In the UK, the mentality is that rich people don’t deserve their wealth and should be taxed into oblivion. The tax authority hunts down any remaining sign of economic dynamism and kills it, then wonders why they can’t build nuclear power plants or maintain steel mills.
Government spending hurts the economy in three ways: (1) inflation is a tax that decreases everyone’s spending power, (2) public sector employment locks talented people out of the productive private economy, (3) government purchasing distorts markets — most prominently in housing, where demand subsidies and debt liberalization have made housing unaffordable, which is a major driver of low birth rates.
China’s looming demographic catastrophe
China has almost certainly overestimated its population by 150 million people. Provincial funding is allocated by population, creating incentive to overcount. Cross-referencing births against vaccines administered reveals overcounting by 220% in some provinces. Most overcounting is in younger age groups.
This is far worse than Japan’s demographic situation. The dependency ratio — old non-productive people supported by young productive people — is becoming catastrophic.
UK is already in this situation: ~3-4% net growth per year is entirely consumed by social spending, so per capita GDP is falling.
Possible solutions: hard cap on elderly spending growth, pandemic that kills old people (kicking the can down the road), AI-powered economic growth exceeding dependency growth, or anti-aging drugs like Ozempic (which already slows aging by 5-10%) scaled up by 10x.
Ultimate solution: longevity escape velocity — slow aging enough that you have time to develop cell rejuvenation, then age in reverse. Like elves in Lord of the Rings who have children rarely but live 10,000 years.
Creating incentives to have more kids
Casey has four children. He used to think you could pay mothers to have more kids, but the real issue is regulatory friction.
Car seat requirements cause many people to stop at two children because fitting three car seats requires an expensive larger vehicle. Car seats save ~100 children’s lives per year but prevent tens of thousands of births — a massive societal own goal.
In some US areas, an unsupervised 17-year-old in public is a crime committed by their guardian. Casey’s almost-7 and 5-year-old can’t walk to school together. When he was 6, he walked alone through forest trails to the ocean to fish and swim.
Jeff Bezos’s mom said she’d rather her kids were missing a finger than lacked access to resources. Casey feels the same — his kids use power tools under supervision and learned to solder at ages 4 and 6.
The best policy for increasing birth rates: ask parents of 3+ kids what it would take to have one more, and remove those barriers. Don’t try to convince childless 38-year-old career women to have one kid — that’s economically negative.
Chewing glass for the love of it
Casey deliberately made himself miserable in high school to build resilience: sleeping outside in the Australian winter with no sheets, just a dew-soaked t-shirt, shivering all night, waking up with spiders on his face — then taking a practice exam to prove he could perform under any conditions.
Running a company involves doing unpleasant things — firing people you love and respect, often because of your own failures. The trick is to rank tasks by unpleasantness and do the worst one first, every day. Eventually you get in the habit of looking forward to it.
Anxiety comes from inaction, not from doing things. Stress and anxiety are nootropics up to a point.
The extent of Casey’s ambition
Casey has come to realize he’s closer to critical infrastructure than he’d hoped. Terraform’s mission is to use cheap solar to produce not just electricity but the entire primary energy consumption layer of civilization:
Desalinating water for Nevada (no big rivers) and the Salton Sea area — turning 110 miles of unwanted coastline into a booming megalopolis with lithium mines, magnesium supersonic aircraft factories, and an inclined plane canal from the Sea of Cortez.
If you want to figure out how to bring titanium to the masses or desalinate 10 million acre-feet of water, Terraform wants to hear from you.
His broader vision: “file print” for hardware. Today you can press print and a document appears. There’s no equivalent for physical objects. Buy a factory. Build 100 factories. The world is waiting.
