YC Application for Deeptech and Hard Tech Startups

Hard tech and deeptech are overlapping terms that YC uses to describe companies where the core challenge is technical rather than market or distribution. Categories include:

• Robotics and automation

• Biotech, synthetic biology, and drug development

• Hardware (medical devices, industrial equipment, consumer hardware)

• Energy (solar, nuclear, battery technology, grid infrastructure)

• Space and aerospace

• Advanced materials and chemistry

• Climate tech with deep technical components

• AI/ML research with novel model architectures

• Quantum computing

What distinguishes these from software companies: the path from idea to working product involves solving a genuinely hard technical problem, not just building software infrastructure. The timelines are longer, the capital requirements are higher, and the evidence available at application stage is fundamentally different.

Because revenue and paying customers are often unavailable at hard tech application stage, YC evaluates a different evidence hierarchy:

1. Technical proof of the core mechanism The most important signal for a hard tech application is proof that the scientific or engineering principle underlying your product works. A published paper demonstrating the mechanism, a lab result proving the core hypothesis, a bench prototype that achieves the target performance metric — these are the equivalent of paying customers for hard tech companies.

State the specific result: "Our prototype achieved 94% accuracy on the target sensing task, compared to the 60% accuracy of current commercial sensors. This performance was validated in 3 independent tests."

2. Working prototype performance against a benchmark For hardware companies, a prototype that performs against a specific technical benchmark is strong evidence. State the benchmark, your current performance, and the gap to commercial viability: "Our current prototype achieves 340Wh/kg energy density. Commercial lithium-ion is at 260Wh/kg. Our target for product launch is 400Wh/kg, which we expect to reach within 8 months based on our current iteration trajectory."

3. Pilot agreements with technical validation partners Letters of intent from companies or institutions willing to pilot your technology — even free of charge — signal that sophisticated buyers find your approach credible. A pilot agreement with a hospital system, a pharmaceutical company, or a large manufacturer is strong validation even without revenue.

4. Team credentials that specifically match the technical challenge For hard tech, who you are matters more than in software because the credibility of your technical approach is inseparable from the credentials of the people pursuing it. A biotech founder with a PhD in the relevant biology from a top institution and previous research in the specific mechanism is more credible than a generalist engineer who has identified the same opportunity.

5. IP position Patents filed or granted, exclusive licenses to foundational research, or novel technical approaches that are not in the public domain are evidence of defensibility that software companies cannot easily claim. If you have IP, mention it specifically.

Each hard tech category has its own evidence milestones. Here are the most relevant by sector:

Robotics:

• Hardware prototype that completes the target task at target speed/accuracy

• Testing results in the target environment (factory floor, agricultural field, hospital room)

• Cost-per-unit estimate at manufacturing scale vs current cost

Biotech / Drug development:

• In vitro or in vivo efficacy data

• Safety profile at relevant doses

• Regulatory pathway clarity (IND application filed, breakthrough therapy designation, etc.)

Medical devices:

• Bench testing results against FDA performance standards

• IRB approval for human trials if applicable

• 510(k) pathway identified or FDA pre-submission meeting held

Energy:

• Energy density, efficiency, or cost metrics at prototype stage vs commercial target

• Cycle life data for battery or storage technologies

• Levelized cost of energy (LCOE) trajectory vs grid parity target

Climate tech:

• Carbon abatement cost per ton vs current alternatives

• Pilot deployment results with specific emissions reduction data

• Comparison to existing technology performance and cost

Advanced materials:

• Material property measurements at lab scale

• Manufacturability evidence — can it be produced at scale?

• Performance comparison to incumbent materials

For each sector, the application should state: current performance, target performance, timeline to target, and the specific technical milestones on the path there.

Hard tech YC companies share consistent founding team profiles. Understanding this profile helps frame your own credentials correctly.

Domain-specific technical depth over breadth. A PhD in the exact relevant subdiscipline is more credible than broad engineering experience. "I have a PhD in solid-state battery chemistry from IIT Bombay and spent 3 years at Tata Chemicals researching lithium-ion alternatives" is more fundable than "I have 10 years of materials science experience."

Direct origination of the technical insight. The most fundable hard tech founders discovered their core technical insight through direct research — they did not read about it. If your company is based on your own research finding, say so explicitly and cite the result.

Commercial awareness paired with technical depth. YC funds hard tech founders who understand both the technical roadmap and the commercial path to deployment. Pure research labs without a commercialization plan are less fundable than research labs with a specific customer and a specific first application. Name your first customer use case and why it is the right beachhead.

A team that covers both technical and commercial depth. One cofounder who is the domain technical expert and one who has commercial experience in the target industry is the ideal hard tech founding pair. A team of two PhDs with no commercial experience raises questions about customer development and distribution. Address any commercial gaps explicitly.

For software companies, the insight is usually about the user — something about your specific buyer that competitors have missed. For hard tech companies, the insight is usually about the technical approach — something about the physics, biology, or engineering that makes your approach work when others have failed.

Strong hard tech insight: "Every solid-state battery company is trying to solve the dendrite problem at high charging speeds. We discovered in our lab research that controlling electrolyte grain boundary orientation during manufacturing eliminates dendrite formation at 5C charging rates. No other team is pursuing this manufacturing-based approach — everyone is pursuing electrolyte chemistry modifications that address the symptom rather than the cause."

This insight is: specific, based on direct research, explains why others have failed, and describes your differentiated approach. That is the shape of a fundable hard tech insight.

Mistake 1: Describing the vision without the technical evidence "We are building the world's most efficient solar cell" is a vision statement, not a technical claim. "Our perovskite-silicon tandem cell achieved 34.1% efficiency in independent testing at IIT Madras in August 2024, compared to the current commercial record of 29.8%" is a technical claim with evidence.

Mistake 2: Oversimplifying the technical novelty Hard tech founders sometimes simplify their technical description to be more accessible. The risk: it makes the application sound like a software company with hardware features. Be specific about what is technically novel. YC partners reading hard tech applications are sophisticated — they will not be confused by technical language.

Mistake 3: No commercial pathway The most common failure in hard tech YC applications is strong technical evidence paired with no commercial clarity. Who is your first customer? What problem does your technology solve for them? What do they currently use instead? What would they pay? Hard tech applications that cannot answer these questions are research projects, not startups.

Mistake 4: Timeline optimism without milestone evidence "We will reach commercial scale in 18 months" without milestone evidence is not credible. "We have reached bench-scale performance at the target specification. Pilot manufacturing is planned for Q2 2025 with [named partner]. Commercial launch is targeted for Q4 2025 based on our current scale-up trajectory" is credible because it names specific milestones with specific evidence.