Deep technology is no longer attracting investment simply because it is innovative. Venture capital is increasingly flowing toward technologies that combine meaningful scientific or technical breakthroughs with clear commercial demand.
That shift was the focus of a recent Entrepreneur Asia Pacific article featuring Senthil M. Kumar, CTO, board member, and senior advisor at Slate Technologies and Celesta Capital.
Drawing on his experience across technology development and venture investing, Senthil explained how the deep-tech market is moving away from experimental concepts and toward technologies that are becoming essential to how modern industries operate.
Venture Capital Is Prioritizing Commercial Readiness
For years, deep-tech investment was often driven by the promise of breakthrough science. Today, technical novelty alone is no longer enough.
Senthil describes this change as capital moving from what is merely “interesting” to what is becoming “inevitable.” The strongest opportunities are emerging where technical maturity, market demand, and commercial viability converge.
According to Senthil, five areas currently demonstrate this shift:
- AI infrastructure
- Industry-specific, or verticalized, AI
- Custom edge silicon
- Supply chain intelligence
- Hard-science climate technologies
Across each category, the most investable companies are not necessarily those producing the most impressive demonstrations. They are the companies successfully turning technical breakthroughs into repeatable, scalable commercial outcomes.
Why Verticalized AI Matters for Construction
One of the most relevant areas for the construction industry is verticalized AI.
General-purpose AI tools can perform well across broad tasks, but they often struggle in complex physical industries where decisions depend on specialized workflows, fragmented project data, regulatory requirements, and real-world operating conditions.
Construction is a clear example.
Projects generate enormous volumes of information across design, planning, scheduling, cost management, field operations, and project controls. However, that information often remains disconnected across systems and teams.
Industry-specific AI can help close that gap by applying construction context to project data, surfacing meaningful relationships, identifying emerging risks, and helping teams make better decisions earlier.
This is also where deep tech becomes commercially relevant. The value does not come from applying AI for its own sake. It comes from helping project teams reduce uncertainty, improve predictability, and turn disconnected information into actionable construction intelligence.
AI’s Next Challenge Is Infrastructure
While much of the AI conversation focuses on models and applications, Senthil argues that some of the most important constraints sit beneath the software layer.
Power, memory, networking, and orchestration software will increasingly determine how reliably and affordably AI can operate at scale.
As AI systems support more users, process larger workloads, and move closer to real-time decision-making, the economics of inference become increasingly important. Energy consumption, memory bandwidth, interconnect speeds, and scheduling efficiency can all limit performance and scalability.
This means the next generation of influential AI companies may not only be building better models. They may be creating the infrastructure required to make AI practical, reliable, and economically sustainable.
For industries such as construction, energy, manufacturing, and utilities, this infrastructure will play an important role in bringing AI into real operating environments rather than limiting it to controlled demonstrations.
Deep Tech Requires More Than a Strong Prototype
The article also explores why many hardware and semiconductor startups struggle despite having impressive technology.
Senthil emphasizes that hardware businesses must coordinate far more than product development. Success also depends on manufacturability, supply chain execution, customer confidence, capital availability, quality control, and market timing.
A prototype may prove that a technology is possible, but it does not prove that the company can produce, deliver, and support it consistently.
That distinction also applies more broadly across industrial technology. Organizations adopting AI and other emerging technologies need confidence that solutions can operate within existing systems, scale across projects, and produce reliable outcomes under real-world conditions.
Commercial readiness requires more than innovation. It requires repeatability, integration, resilience, and trust.
How Investors Evaluate Deep-Tech Companies
Deep-tech investors assess risk differently than investors evaluating traditional software businesses.
In many software companies, the central questions involve adoption, retention, and growth. In deep tech, investors must also determine whether the technology can be built, manufactured, qualified, deployed, and delivered at a commercially viable cost.
As a result, deep-tech due diligence evaluates several interconnected factors:
- Scientific and technical validity
- Engineering execution
- Capital requirements
- Manufacturing and supply chain readiness
- Market timing
- Ecosystem dependencies
- The team’s ability to translate technical expertise into business value
Senthil describes the strongest deep-tech teams as bilingual. They understand both the underlying science and the commercial realities required to build a sustainable business.
Resilience and Trust Are Becoming Competitive Advantages
Geopolitical risk and supply chain concentration are also changing how semiconductor and hardware companies approach growth.
Manufacturing capacity is concentrated across a limited number of regions, and production cannot always be shifted easily from one facility or supplier to another. Startups must therefore consider resilience, compliance, sourcing, and geographic exposure earlier in their development.
Senthil argues that the strongest companies will build this resilience into their strategies from the beginning. They will diversify key partnerships, account for regulatory requirements, and avoid relying on a single supplier or geopolitical assumption.
In this environment, trust becomes a competitive advantage.
Customers, partners, and investors need confidence that a technology company can deliver consistently, respond to disruption, and continue operating as market conditions change.
What This Shift Means for the Built Environment
The movement from experimental deep tech to commercially necessary technology has major implications for construction and infrastructure.
AI is becoming increasingly embedded in how organizations plan projects, manage information, evaluate risk, and make decisions. However, the technologies that create lasting value will be those designed around the realities of the industries they serve.
For construction organizations, that means moving beyond disconnected tools and isolated AI experiments. The opportunity is to create a connected intelligence layer that understands project context, brings together information across the project lifecycle, and helps teams identify risks and opportunities sooner.
At Slate Technologies, we are focused on applying industry-specific AI to the complexity of construction and capital projects. Slate’s AI-powered construction intelligence platform connects project data and transforms it into insights that support more informed, predictable project delivery.
As deep tech continues moving from interesting to inevitable, the organizations that benefit most will be those that connect technical innovation to practical, repeatable business value.