Vijay Pande spent years managing roughly $4 billion in biotech investments at Andreessen Horowitz. Now he's running a smaller, AI-native fund called VZVC—and making drastically fewer bets per year. His reasoning? The game has changed. Biology is transitioning from a discovery science to an engineering discipline, and investors are getting pickier about what deserves capital.
For founders building at the intersection of AI and biology or health, Pande's shift carries a clear message: in 2026, you need a real product and early traction to stand out. The era of funding thirty moonshot research projects annually is over. The new era rewards engineering rigor, repeatable outcomes, and proof that your approach actually works.
From Discovery to Engineering: What Changed
Pande's central thesis is that biology has fundamentally shifted. Thanks to advances in AI, computational tools, and data infrastructure, we're no longer limited to trial-and-error discovery. Instead, teams can engineer predictable, reproducible outcomes—the same way software engineers ship reliable products.
This evolution mirrors what happened in software over the past two decades. Early internet startups often raised millions on an idea and a slide deck. Today's founders are expected to ship working code, demonstrate user traction, and prove unit economics before institutional capital flows in. Biology and health tech are entering that same maturity curve.
The implication for founders is straightforward: if you're building in this space, show the engineering. Demonstrate that your platform or product produces consistent results, not one-off breakthroughs that may or may not replicate. Investors like Pande are looking for systematic, repeatable processes that de-risk the notoriously uncertain path from lab to clinic.
Open Data Wins Over Walled Gardens
Pande also emphasizes that open, shared datasets—not proprietary walled-off data—are what will enable AI to truly transform medicine. Clinical trials remain brutally expensive, often running into hundreds of millions of dollars. Reducing that cost and timeline requires collaboration, shared infrastructure, and access to large-scale, high-quality data.
For early-stage founders, this is both a strategic opportunity and a practical filter. If your approach depends on hoarding data or building everything in-house, you may struggle to attract mission-aligned, sophisticated investors. Conversely, if you can show how your product leverages open datasets, public research, or collaborative platforms to accelerate progress and reduce risk, you'll stand out.
This also speaks to a broader trend: the most compelling AI-driven health companies aren't trying to reinvent the entire stack. They're building on shared foundations, contributing back to the ecosystem, and proving value in tightly scoped use cases before expanding.
Fewer Bets, Higher Standards
The shift from "30 bets a year" to a concentrated portfolio reflects a broader recalibration in venture capital. When money was cheap and every pitch deck promised a billion-dollar outcome, investors could afford to spray capital across dozens of speculative projects. In 2026, capital is more expensive, exit timelines are longer, and diligence is deeper.
Founders need to adapt. That means:
- Shipping working prototypes that prove the core engineering thesis, not research projects that promise future breakthroughs.
- Demonstrating traction with real users, customers, or clinical partners—even at small scale.
- Showing a realistic path through regulatory and clinical hurdles, with credible cost and timeline assumptions.
- Positioning your product as infrastructure-aware, leveraging shared datasets or platforms to reduce risk and accelerate time-to-value.
In short, you need to act less like a research lab and more like a disciplined product team.
Key Takeaways
- Vijay Pande, former a16z biotech lead, is now making fewer, more concentrated bets through his new AI-native fund VZVC.
- Biology is shifting from discovery to engineering—investors want repeatable, predictable outcomes, not one-off breakthroughs.
- Open, shared datasets will drive AI transformation in medicine more effectively than proprietary walled gardens.
- Clinical and regulatory costs remain high; show a realistic, capital-efficient path to validation.
- In 2026, VCs expect working products and early traction, not just compelling pitches.
Why This Matters for Your MVP
If you're building at the intersection of AI, biology, or health, the bar for institutional funding has risen. But that's exactly why a working, sellable MVP is your strongest asset. A disciplined product that proves the engineering thesis, runs on real data, and shows measurable traction will separate you from the crowd of pitch decks and research proposals.
Speed to working product is now a competitive advantage—not just in software, but across every domain where AI is enabling engineering-level rigor.