The hook is a macro event that most crypto investors have missed. In late 2024, Cadence CEO Anirudh Devgan told a tech publication that his company is “undervalued” amid the AI boom. The statement landed with a thud in traditional finance, but it should have echoed louder in crypto. Because the same AI infrastructure that is driving Nvidia’s trillion-dollar market cap is also the backbone of the next generation of crypto mining, AI inference chips, and decentralized compute networks. Cadence is the silent enabler — the picks-and-shovels vendor that every chip designer, from Nvidia to Bitmain to a team building a custom ASIC for a Layer 1 blockchain, must pay. And the market is pricing it as if it were just another cyclical software company.
Context: What is EDA and why should crypto care?
Electronic Design Automation (EDA) is the software used to design and verify chips. Without EDA, no modern chip — whether a GPU for AI training, an ASIC for Bitcoin mining, or a custom accelerator for a decentralized AI network — can be built. Cadence and Synopsys together control over 60% of the global EDA market. Their tools are the operating system of chip design.
Crypto native projects often overlook this layer. Many focus on the token economics of DePIN networks or the hashrate of Bitcoin mining, but the underlying hardware design chain is where the real structural leverage sits. Each new generation of mining ASIC requires EDA tools that cost tens of millions of dollars in licensing fees. Each custom AI chip for a project like Bittensor or Akash Network requires the same. The more chips the world designs, the more Cadence earns.
Cadence’s revenue model is shifting from traditional perpetual licenses to subscription-based, with a growing component of usage-based pricing tied to the number of design starts. This means its revenue is becoming more recurring and more correlated with the global chip design activity — which is exploding thanks to AI and, to a lesser extent, crypto’s hardware demands.
Core Analysis: The Structural Undervaluation of Cadence
1. Technical Position and Moat Cadence’s core product portfolio covers three domains: EDA toolchain, functional verification and IP, and system analysis (multiphysics simulation). Its process node support spans from mature 28nm to bleeding-edge 3nm/2nm GAA. The moat is threefold: decades of accumulated design IP and process PDK databases, deep co-optimization with foundries (TSMC, Samsung, Intel), and full-flow consistency from RTL to GDSII.
For crypto-specific chips, the requirements are often extreme. Bitcoin mining ASICs push the boundaries of low-power, high-throughput design. AI inference chips for decentralized networks demand energy efficiency and custom acceleration. Cadence’s tools are the industry standard for both.
2. AI-EDA Integration Cadence has launched the Cadence.AI platform, embedding machine learning into the design flow to automate optimization. This is a double exposure: Cadence both enables AI chip design and uses AI itself. CEO Devgan’s point is that the market underweights the latter. The AI-EDA synergy is a growth multiplier that traditional valuation models fail to capture.
Based on my audit experience dating back to 2017, I have seen how software toolchains evolve: the ones that embed AI into their own workflows tend to widen their moat faster than those that simply sell to AI customers. Cadence is doing both.
3. The Leverage Ratio The EDA industry is a small fraction of the global semiconductor market — about $150-180 billion in 2024, less than 3% of the $600 billion semiconductor industry. But its leverage effect is massive: every $1 of EDA revenue supports an estimated $200-300 of semiconductor output and $5,000-10,000 of end-user tech value. This “superlinear leverage” is a systematic reason why Cadence is undervalued. The market sees a $40 billion revenue company, not the $1 trillion+ ecosystem it enables.
4. Demand Drivers from Crypto Crypto-specific demand is often underestimated. The global mining ASIC market alone is over $5 billion annually, and each new generation of ASIC requires EDA tools. More importantly, the rise of AI+blockchain projects — such as Bittensor, which uses custom hardware for training, or decentralized physical infrastructure networks (DePIN) that rely on specialized chips — is creating a new class of chip design customers. These customers are not yet material for Cadence, but the trend is accelerating.
5. Hidden Information: The License-to-Subscription Migration The CEO’s undervaluation claim hints at a deeper structural shift: EDA is moving from a software licensing model to a platform-based subscription model, and eventually to a revenue-sharing model tied to the chips’ success. If Cadence begins to take a percentage of chip revenue (as opposed to a flat software fee), its profit elasticity would be radically higher. This is the “AI infrastructure tax” narrative. The market has not priced this optionality.
Historically, Synopsys and Cadence have traded at multiples similar to other software companies. But the nature of their business — essential, non-discretionary, and growing with chip complexity — warrants a premium. The fact that the market still lumps them into “application software” is a valuation anomaly.
Contrarian Angle: The Decoupling Thesis is Wrong for EDA
The conventional wisdom in crypto is that the sector is decoupling from traditional tech. But that is a surface-level view. The hardware underpinning both AI and crypto mining is converging. The same TSMC process that makes Nvidia’s H100 also makes Bitmain’s Antminers. The same EDA tools that design an AI GPU also design a blockchain oracle chip.
The contrarian take is that Cadence (and Synopsys) are actual beneficiaries of the AI boom that crypto investors ignore. They are the ultimate “shovel sellers” in a gold rush that spans both AI and crypto. The market’s failure to price this correctly is a structural blind spot, not a temporary mispricing.
*
Takeaway: Positioning for the Cycle
Crypto investors should watch the EDA sector as a proxy for the health of the entire hardware design ecosystem. When Cadence’s CEO says the company is undervalued, he is not just pitching a stock. He is signaling that the market is missing the leverage of the AI infrastructure wave. For those who can see the pattern, the question is not whether to buy Cadence, but whether the next cycle will reward those who saw the structural integrity of the chip design supply chain before the sentiment caught up.
History repeats not in price, but in pattern. The pattern here is clear: every technological revolution that requires custom silicon — from the internet to AI to crypto — first enriches the toolmakers. The market has not yet priced that for Cadence.
Logic is immutable; incentives are the variable. The audit passed, but the economics failed. Structural integrity precedes market sentiment.