Delian Asparouhov on holiday term sheets, defense manufacturing unicorns, and the K-shaped LP market

Summary

Delian Asparouhov, co-founder of Varda and partner at Founders Fund, closed a Series C term sheet over the 2025 holiday period, extending a personal pattern he describes as signing a term sheet within 48 hours of Christmas Day in seven of his nine years in venture. The holiday window works, he argues, because founders can test financing conversations quietly, away from board and employee scrutiny, and investors can enforce hard deadlines — sign by New Year's or the deal dies — that would be socially difficult to impose during normal operating periods.

Barbell Strategy in Deep Tech

Asparouhov has structured his investing around a barbell that skips what he calls the "messy middle." He enters at the pre-seed or seed stage in industrial deep tech companies, then waits — sometimes years — for the specific inflection point just before economics turn, and re-enters with a growth-stage check. Over the past year he completed three such later-stage rounds at Series B, Series C, and a second Series C closed over the holiday period. All three were companies he had tracked or backed since seed.

The rationale is grounded in how price-per-share curves behave in hardware versus software. Deep tech companies typically see a 4x to 5x step up from seed to Series A, then years of flat or minimal PPS appreciation while grinding through R&D and production hell, followed by sharp step-function re-ratings. That profile, he argues, is fundamentally different from SaaS or AI companies like Ramp, which show smoother, more continuous compounding.

Hre: Defense Manufacturing as a Neo-Prime

The clearest illustration of the strategy is Hre (referred to as "Adrian" at points in the conversation), an automated CNC metal-cutting business that serves aerospace and defense customers including rocket and drone manufacturers in the Founders Fund portfolio. The company just closed a $1.6 billion post-money round led by T. Rowe Price, which Asparouhov describes as roughly his fifth unicorn outcome from a sub-25 seed investment.

Hre's positioning is distinct from other defense neo-primes like Anduril and Shield AI, which are primarily product developers that outsource subcomponent manufacturing. Hre is explicitly a production-focused neo-prime — it amplifies others' end products rather than designing its own. That makes it a direct answer to the Pentagon's manufacturing bottleneck, winning contracts both as a standalone prime and as a co-bidder alongside other neo-primes.

The investment thesis crystallized in Q1 2025 when two things converged: the unit economics finally turned positive after years of production grind, and it became clear the defense industry's shake-up would elevate Hre from backend vendor to strategic prime. Asparouhov's read at the time was that margins would compound — from breakeven toward 20%, 40%, then 60% — and that the structural demand for machined metal parts is closer to semi-infinite than the addressable market for most enterprise software. The reason Hre chose T. Rowe Price over traditional Silicon Valley venture firms, he notes, is straightforward: T. Rowe is likely to underwrite the IPO and hold the stock post-listing, giving the company a long-term institutional anchor that a typical growth fund cannot replicate.

K-Shaped LP Market

The broader venture fundraising environment is bifurcating sharply. Total dollars raised by venture firms through traditional fund structures has declined linearly since the 2021-22 peak — a trend visible in Carta data — while total LP capital flowing into tech via direct investments, SPVs, and co-investment vehicles has almost certainly increased over the same period.

The dynamic creating pressure on mid-tier and emerging managers is structural. When a handful of mega-rounds — effectively a "private Mag 7" — dominate return expectations, LPs can access those same logos through fee-light or fee-free SPVs from lesser-known managers, bypassing the traditional 2-and-20 structure. Top-tier funds with strong markups are still struggling to close fully because LPs are weighing a marked-up fund position against direct SPV access to the same underlying companies. Emerging managers face a compounding problem: fewer new managers are launching, close timelines are extending, and even established multi-stage firms are finding it harder to justify raising large growth vehicles when the assets those funds would target are already accessible directly.

For endowments specifically, the public market rally in Mag 7 names has created an inverse version of the 2020 problem. In 2020, private books weren't marking down fast enough, pushing private allocations to 20% of endowment portfolios and forcing investment freezes. Now some endowments that target a 7% private allocation have drifted to roughly 3.5% and are actively looking to deploy — but they are arriving with large checks and a short list of names they want, making direct access to those names increasingly attractive relative to fund intermediation.

Asparouhov expects fee compression to follow, particularly for late-stage venture funds, and predicts liquidity preferences — a feature of private rounds that public investors at equivalent valuations never receive — will become increasingly difficult to justify and are likely to erode. The 2021 experiment of VC firms trying to operate like crossover funds managing public equity books, he says, was broadly a failure, and most have retreated from that model.