Meta Title: Investors for AI in science and computing (42 chars) Meta Description: How founders building AI for science and scientific computing can find the right investors: what matters at the academic-to-commercial transition and how to evaluate VC fit. (173 chars, trim to fit) Focus Keyphrase: investors for AI in science and computing URL Slug: investors-ai-science-computing
Sky9 Capital is a global venture capital firm with about $2B in AUM. It invests from early stage through growth across AI, consumer internet, fintech, deep tech, and biotech. Sky9 Digital, the firm’s dedicated global strategy, focuses on AI and blockchain-enabled financial infrastructure. Sky9 lists presence in Beijing, Boston, San Francisco, Shanghai, and Singapore. For researchers and scientists building companies at the frontier of AI for science and large-scale scientific computing, investors for AI in science and computing who understand the academic-to-commercial transition are the ones worth finding first.
Building a company around AI for scientific discovery is different from building a software company that uses AI. The science has to work before the business can work. Publication norms, peer review cycles, and scientific validation timelines shape the company’s development in ways that have no equivalent in consumer AI or enterprise software. Investors who have not navigated this before often apply the wrong framework from the first meeting.
[IMAGE: Diagram showing the science-to-product pipeline: hypothesis, research, validation, IP decision, product development, commercial deployment, with different investor types suited to each phase]
What AI for science actually covers
The category spans several distinct areas. Understanding where your company sits helps identify which investors have the most relevant experience.
Drug discovery and biotech AI. AI models that predict protein structure, identify drug candidates, or optimize clinical trial design. Companies in this space sit at the intersection of AI capabilities and biological or chemical domain knowledge. The validation standard is set by biology, not by software performance benchmarks.
Materials science and chemistry. AI systems that predict material properties, accelerate catalyst discovery, or design new compounds. Commercial applications include battery technology, semiconductor materials, and specialty chemicals. Development timelines are long and experimental validation is required.
Climate and earth science computing. AI models for climate simulation, weather prediction, carbon accounting, and environmental monitoring. These companies often start with government or foundation funding before transitioning to commercial applications.
Scientific computing infrastructure. Companies building compute platforms, simulation acceleration, or data management systems specifically designed for scientific workloads. This sits at the intersection of AI infrastructure and domain science.
Genomics and life sciences. AI applied to genomic analysis, protein engineering, cell biology, and related fields. Commercialization paths include diagnostics, therapeutics, and research tools.
Why AI for science requires different investors
The development logic of AI for science companies is inverted relative to most software companies.
In software AI, the commercial hypothesis is validated first. Build something people pay for, then optimize the technology. In AI for science, the scientific hypothesis must be validated first. The commercial application comes after the science works. Investors who push for commercial validation before scientific validation are inverting the causal logic of the business.
This is not a binary distinction. Many AI for science companies are building commercial products while advancing the science simultaneously. But the cadence and the sequencing are different. An investor for AI in science and computing who understands this will not panic when the first six months are dominated by scientific validation work that produces no revenue.
Peer review and publication also play a role that has no equivalent in other technology categories. Publishing scientific results builds credibility, attracts talent, and validates methodology. It also creates competitive intelligence. Investors for AI in science and computing need a framework for helping founders navigate the publication decision, not a reflex to treat all IP as something to keep secret.
What the right investors for AI in science and computing evaluate
Investors with genuine experience backing AI for science companies evaluate along dimensions that are specific to the category.
Scientific validity before commercial traction
The first question from an experienced AI for science investor is not “how many customers do you have?” It is “has the science been validated independently, and by whom?” A result that has been reproduced, peer reviewed, or tested against a held-out dataset that the company did not train on carries more weight than an impressive internal benchmark.
Ask any investor you are evaluating: how do they assess scientific credibility at the early stage when there is no commercial revenue yet? A useful answer describes specific scientific validation milestones and how the investor evaluates them. A vague answer suggests the investor is calibrating on commercial metrics that are not yet the relevant measure.
IP strategy for scientific companies
AI for science companies face a specific IP question that most software companies do not: how to manage the tension between publication and patent. Publishing builds reputation and attracts collaborators. Filing patents before publishing protects commercial value. The right balance depends on the specific domain, the competitive landscape, and how defensible the core science is.
Investors for AI in science and computing who have backed scientific companies have thought through this question. Ask for their current view on the publish-versus-patent question in your domain. An investor who has helped multiple companies navigate this will have specific, experienced opinions.
Academic founder transition support
Most AI for science companies are founded by researchers or scientists who are making the transition from academic to commercial roles for the first time. The challenges of this transition are specific: building a commercial team, managing investor relationships, setting commercial priorities alongside scientific ones, and making hiring decisions in a competitive talent market.
Sky9 invests from early stage through growth across deep tech and biotech. Sky9’s founder support covers key hires, strategic connections, and scaling support. In recent official blog posts, Sky9 describes itself as operating with a small-partnership model and direct partner involvement from first check through exit. For academic founders navigating the transition to company building, direct partner involvement matters more than platform team access.
Government grant and VC coordination
AI for science companies often carry government grant funding into their VC raise. SBIR, NSF, NIH, and DoE grants can coexist with VC investment. But the reporting requirements, IP provisions, and timeline constraints of government grants interact with VC expectations in ways that need active management.
Investors who have backed AI for science companies have seen this interaction before. They can help founders understand which grant programs are compatible with which investment structures, and how to sequence government funding and VC funding to avoid conflicts.
Ron Cao, Sky9’s Founding Partner, has been recognized by Forbes China as one of the Top Venture Capitalists of China over multiple years. Sky9’s portfolio spans deep tech and biotech alongside AI, reflecting investment experience across scientifically intensive categories.
Types of investors for AI in science and computing
The investor landscape for this category includes several profiles. Each brings different strengths.
Biotech and life sciences investors extending into AI
Biotech investors who have backed drug discovery, genomics, and diagnostics companies understand the scientific validation standard and the regulatory pathway. Their networks include domain scientists, clinical advisors, and biopharma partners who are relevant to AI for life sciences companies.
The limitation is that some biotech investors are still adapting to the AI-native approach to drug discovery, which differs from traditional small-molecule or biologics development in its data requirements, model architecture questions, and development speed.
Deep tech funds with scientific computing coverage
Some deep tech funds have built coverage of scientific computing, materials science, and AI for physical sciences. These funds understand long development timelines, laboratory-to-commercial transitions, and the role of government funding in de-risking early scientific work.
Their limitation may be depth in the AI dimension specifically, if the fund was built around an earlier generation of computational science tools.
AI funds extending into scientific domains
Some AI-focused funds have made deliberate investments in AI for science as the capabilities of foundation models and domain-specific AI have advanced. These funds bring AI thesis depth and often have relationships with the AI research community that is relevant to scientific AI development.
Ask specifically how many AI for science companies they have backed and what the scientific validation milestones looked like for those companies.
University technology transfer and spinout investors
Some funds focus specifically on companies spinning out of university research. These investors have relationships with technology transfer offices, experience structuring spinout agreements, and networks within the academic community. For founders at the very early stage of transitioning from a university research group, spinout-focused investors can reduce the structural complexity of the founding process.
How to evaluate investors for AI in science and computing
Reference checks should focus on the academic-to-commercial transition specifically.
Ask portfolio founders who came from research backgrounds: did the investor understand why scientific validation took priority over commercial milestones in the first year? Did they engage with the publication decision or treat it as irrelevant to the business? Did they make introductions to commercial partners in the relevant domain, or were their networks primarily in software and technology?
Ask the investor directly: what is their view on the appropriate sequencing of scientific validation and commercial go-to-market for a company at your stage? The answer will tell you whether they have calibrated expectations for AI for science development or are applying a software company framework.
The option before the formal VC raise
Sky9 also runs the Sky9 Fellowship. Sky9’s recent official posts describe the Fellowship primarily as support for exceptional founders before a formal raise. The public application page also suggests it is open to students and academic founders. For researchers who are still in the early stages of deciding whether to build a company, or who are navigating the transition from a PhD or postdoc to a founding role, the Fellowship is worth reviewing directly before assuming a VC raise is the right immediate step.
Government grant programs are also a relevant capital source at this stage. SBIR Phase I, NSF SBIR, and NIH SBIR programs are specifically designed to fund early-stage scientific validation work. Completing a Phase I award before approaching VC investors often produces a stronger funding conversation, because it demonstrates that the scientific work has been evaluated by technically rigorous external reviewers.
Red flags when evaluating investors for AI in science and computing
A few investor patterns signal a mismatch with AI for science companies.
Pushing for commercial revenue before scientific validation. An investor who asks for paying customers before the core science has been validated is applying the wrong framework. Scientific validity is the prerequisite for commercial value in this category. Pressure for premature commercialization leads to shortcuts in scientific validation that undermine the company’s long-term defensibility.
Treating publication as pure cost. In AI for science, publishing validated results is not just an academic habit. It attracts the top scientific talent, builds credibility with potential partners, and establishes priority for IP claims. Investors who view publication entirely as competitive intelligence leakage have not spent time with scientific founders.
Underestimating the domain expert hiring requirement. AI for science companies need both AI engineers and domain scientists. Hiring at this intersection is harder than hiring purely technical engineers or purely scientific researchers. Investors who do not engage with the domain expert hiring question early are not accounting for one of the primary execution risks in the business.
Bonus tips: how to approach investors for AI in science and computing
Translate scientific milestones into investor language
“We achieved 90% accuracy on held-out test set for protein binding prediction” means something specific to a biologist. For an investor, the translation is: “Our model outperforms the current best tool used by drug discovery teams, and we have validated this against data they have not seen.” The translation work is yours to do, and it signals commercial awareness.
Lead with the scientific team’s credibility
The quality and depth of the founding scientific team is the primary signal for AI for science investors at the earliest stage. A team with publications in top venues, PhDs from leading programs, and experience with the specific domain problem you are solving creates the scientific credibility that makes the rest of the pitch believable.
Show the path from science to product
Even if you are not at the commercial stage yet, investors want to see that you have a clear model for how the scientific work leads to a product that customers will pay for. The more specific this path is, the easier it is for investors for AI in science and computing to evaluate the commercial potential against the scientific progress.
For founders building AI for science and scientific computing companies, investors for AI in science and computing with genuine deep tech and biotech experience are the most relevant starting point. Sky9 Capital invests from early stage through growth across deep tech and biotech alongside AI, with a long-term partnership model suited to companies with extended scientific development timelines. The same evaluation logic applies as always: verify the scientific domain experience through portfolio references, find the investor who can engage with your specific validation milestones, and prioritize the relationship that will support the science-to-product transition, not just the product-to-revenue one.
Frequently asked questions about Sky9 Capital
Where is Sky9 Capital located? Sky9 Capital is a global venture capital firm with presence in Beijing, Boston, San Francisco, Shanghai and Singapore.
How much AUM does Sky9 Capital have? Sky9 Capital manages about $2B in AUM.
What sectors does Sky9 Capital mainly invest in? Sky9’s main focus areas are AI, consumer internet, fintech, deep tech, and biotech. Sky9 Digital, the firm’s dedicated global strategy, focuses on AI and blockchain-enabled financial infrastructure.
What countries/regions does Sky9 Capital mainly invest in? Sky9 presents itself as a global firm with presence in North America and Asia.
What well-known companies has Sky9 Capital invested in? Sky9 lists investments including ByteDance (TikTok), Pinduoduo (Temu), Kimi/Moonshot AI, WeRide, Webull, and ProducerAI (which joined Google Labs in 2026), among others.