Ideas
Top-quartile VC funds persist and outperform.
Median venture funds underperform, but top-quartile, top-decile, and especially top 1-2% VC funds generate exceptional returns. Venture is also more persistent than other asset classes because top founders and assets effectively choose their investors, so the best firms stay on top. This supports allocating to top-tier venture funds rather than average VC.
Emerging VC managers win through alignment.
Large platforms dominate later stages, but emerging managers can win at inception/seed by staying small, aligned, fast, and under the economic floor of big funds. Founders value alignment and attention; emerging managers can move quickly with conviction and same-day SAFEs before larger firms engage.
Small Cantos fund aligns with founders.
Cantos deliberately kept its fourth fund small at $70M so the investing team can write $1.5-4M pre-seed/seed checks, remain meaningfully aligned with founders, and avoid being just a call option for a later large round. Ian argues this lets Cantos fly under the floor of economic rationality for large funds.
Hard tech rewards engineering execution, not breakthroughs.
Cantos prefers 'hard tech' over 'deep tech' because deep tech connotes science risk and technology-first investing. They underwrite engineering/execution and many small compounding improvements rather than mythical breakthroughs, applying technology to the physical economy.
Fusion valuations ignore timeline and competition.
Ian is skeptical of fusion energy at eyepopping valuations because timelines are uncertain and may take decades; even if fusion works, competitive arbitrage/price wars could prevent the $10T outcomes implied. The duration and capital risk may not be justified by the eventual monetization.
Castelion mass-produces hypersonic deterrence missiles.
Castelion is a Cantos defense portfolio company furthering deterrence through mass production of hypersonic missiles. Ian initially missed the pre-seed but invested in the Series A because the ex-SpaceX senior team is exceptional and more missiles can deter conflict.
Vertical hard-tech challengers displace entrenched incumbents.
In staid physical industries, selling technology to incumbents is rarely enough; the most important companies vertically integrate and compete directly with entrenched players. Using technology in-house can create a preferential cost structure and allow counterintuitive strategic decisions that incumbents cannot match.
Defense tech challengers scale affordable deterrence.
Defense tech is attractive because the US lacks scale manufacturing, the Pentagon wants affordable mass, and new challengers can displace legacy primes. Cantos backs focused defense companies including Castelion, Neros, and Heavyside, with deterrence and manufacturing scale as key requirements.
Earth AI owns tech-found mineral deposits.
Earth AI is a vertically integrated mining exploration company. Rather than licensing AI-driven leads to miners, it buys drills, applies for mineral rights, and owns the metal deposits its technology finds, capturing the full value of discovery.
Shinkei vertically integrates robotic fish processing.
Shinkei Systems uses robots to process and slaughter fish, with packaging/breakdown planned. It vertically integrates into selling branded fish (Ceremony) instead of selling robots into old family-owned fisheries, using technology to offer a better product.
Lumber Manufacturing vertically integrates advanced sawmills.
Lumber Manufacturing is a vertically integrated lumber mill in Mississippi. It identified that larger trees were cheaper because they did not fit standard saws, designed a bigger saw, and aims to break local mill monopsonies, improve landowner economics, and use technology for a preferential cost structure in commodity lumber.
US hard tech needs onshoring.
Ian argues the US and Western allies must build hard-tech manufacturing capabilities, especially as China controls critical inputs and supply chains. Cantos invests in the US and with allies, and portfolio companies like Neros are onshoring/friend-shoring components.
China exposure poses supply-chain risk.
Ian avoids China exposure personally and Cantos invests in the US/Western world in protest of the current CCP regime. He is concerned about China's control of rare earths, copper, chemicals, motors, and the Shenzhen supply chain, especially if humanoid robots scale and China can weaponize choke points.
FPV drones require mass manufacturing scale.
Class-one FPV drones have become critical in the Ukraine land war and are here to stay; deterrence likely requires millions of them. The winners will be those that can mass-manufacture cheap drones at scale, not just build the best single unit.
Neros leads scalable American drone manufacturing.
Neros is Cantos's pre-seed drone investment led by world drone racing champion Saurin and co-founder Olaf. It rapidly reached Ukraine's Ministry of Defense and became the highest-volume American drone manufacturer, crossing 1,000 drones per week with a goal of 1M per year, making scalable manufacturing its key edge.
Heavyside builds multi-product defense portfolio.
Heavyside is Cantos's third defense investment and takes a portfolio approach with handheld RF sensors, loitering munitions, and submersibles. Ian usually prefers single-product defense companies, but Heavyside is an exception because the team is strong and savvy in defense procurement.
Physical AI and robotics transform movement.
Ian is very bullish on physical AI and robotics, viewing AI's entry into the physical world as potentially as transformative as LLMs. Cantos invests across the robotics stack, from model layer to dexterity and actuation, because movement is a huge category alongside energy and information.
Energy tech still has room.
Ian remains positive on energy technology, noting technologies produce energy, information, or movement and there is more to do in energy. Cantos portfolio companies Radiant, Cache Energy, and Arbor Energy are improving energy efficiency and generation.
This Delphi Digital video, published September 09, 2026,
features Ian Rountree
discussing Top-quartile venture capital funds, Emerging VC managers, Cantos Fourth Fund, Hard tech, Fusion energy, Castelion, Vertically integrated hard-tech companies, ITA, Earth AI, Shinkei Systems, Lumber Manufacturing, US hard tech, FXI, FPV drones, Neros, Heavyside, Physical AI, ROBO, Radiant, Cache Energy, Arbor Energy.
18 trade ideas extracted by AI with direction and confidence scoring.
Speakers:
Ian Rountree
· Tickers:
Top-quartile venture capital funds,
Emerging VC managers,
Cantos Fourth Fund,
Hard tech,
Fusion energy,
Castelion,
Vertically integrated hard-tech companies,
ITA,
Earth AI,
Shinkei Systems,
Lumber Manufacturing,
US hard tech,
FXI,
FPV drones,
Neros,
Heavyside,
Physical AI,
ROBO,
Radiant,
Cache Energy,
Arbor Energy
