Welcome to today's episode. Deutsche Börse Group closed financial year 2025 with record results across every key metric — the highest net revenue, EBITDA, net income, and earnings per share in the company's history.

Net revenue without treasury result rose 9% to €5.2 billion, while operating costs grew by only 3% — a discipline that drove EBITDA up 14% to €2.7 billion. Net profit reached €1,995 million, with cash EPS at €11.65.

All four business segments contributed: Investment Management Solutions, Trading & Clearing, Fund Services, and Securities Services all posted solid growth — with Fund Services up 24% in EBITDA and Securities Services up 20%.

The Group rewarded shareholders with a 5% dividend increase to €4.20 per share and launched a €500 million share buyback program. Looking ahead, Deutsche Börse targets €5.7 billion in net revenue for 2026.

Today, we deep-dive into what's really driving this machine — and what it means for investors.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Circle Q4 2025 — Earnings Analysis

Circle Internet Group just delivered one of its strongest quarters on record — and the numbers speak for themselves.

In Q4 2025, Circle reported total revenue and reserve income of $770 million, a staggering 77% increase year-over-year. Adjusted EBITDA surged 412% to $167 million, demonstrating the extraordinary operating leverage embedded in their business model. Earnings per share came in at $0.43, crushing the analyst consensus of $0.35 — a beat of nearly 23%. The market reacted accordingly: shares jumped as much as 30% in post-earnings trading.

The engine behind all of this? USDC. Circle's flagship stablecoin ended the year with $75.3 billion in circulation, up 72% from the prior year. Even more telling is the velocity of usage: on-chain USDC transaction volume hit $11.9 trillion in Q4 alone, representing a 247% year-over-year surge. This is not just supply growth — it's adoption growth. More users, more protocols, more transactions.​

For the full year 2025, Circle posted $2.7 billion in total revenue and reserve income, up 64% versus FY2024. The company did report a net loss of $70 million for the year — but context matters: that loss was entirely driven by $424 million in stock-based compensation triggered by vesting conditions tied to the IPO. Strip that out, and the underlying operating performance is solidly profitable.​

The key structural risk remains the same as always: Circle's reserve income is highly sensitive to interest rates. With 95%+ of revenue still tied to yield on USDC reserves, any meaningful rate decline creates headwinds. In Q4, the reserve return rate was already down 68 basis points year-over-year. This is the tension every Circle investor must hold in mind — explosive volume growth on one side, rate sensitivity on the other.​

The bottom line: Circle is executing at a high level. The USDC flywheel is accelerating, margins are expanding, and the business is proving it can scale. The question now is whether the regulatory tailwinds in the U.S. stablecoin landscape — and the IPO momentum — can carry this into 2026.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

NVIDIA Q4 FY2026 — Record Results and Structural Implications

NVIDIA closed fiscal year 2026 with historic results, confirming its position at the center of the global AI infrastructure build-out.

In Q4 alone, revenue reached $68.1 billion, up 73% year over year. Full-year revenue came in at $215.9 billion, up 65%. Net income totaled $120 billion, and diluted EPS rose 67% to $4.90.

Free cash flow was exceptional: $35 billion in the quarter and $97 billion for the full year — implying an FCF margin close to 45%. At this scale, that level of cash generation is extraordinary.

The core driver remains Data Center, which generated $51.3 billion in Q4 revenue. The Blackwell architecture is now the dominant platform, shifting NVIDIA from selling individual GPUs to delivering full AI systems — including compute, networking, and software.

Networking has become a major pillar as well, surpassing $11 billion in quarterly revenue and over $31 billion annually. With Spectrum-X and integrated AI networking, NVIDIA is capturing more of the total data center stack.

Gross margin for the year stood at 71.1%. While the transition to full-system Blackwell solutions introduces operational complexity, it also strengthens pricing power and long-term operating leverage.

Looking ahead, Q1 FY2027 guidance points to approximately $78 billion in revenue, indicating continued sequential growth and strong demand visibility extending into calendar 2027.

Key risks remain: export restrictions to China, hyperscaler CapEx cyclicality, and supply constraints in advanced memory.

Bottom line: NVIDIA is no longer just a chip company. It is the infrastructure backbone of the AI revolution. The strategic question now is not about growth — it is about sustainability and valuation at scale.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Today we’re looking at Applied Materials and its latest quarterly results — a clear reflection of how artificial intelligence is reshaping the semiconductor industry.

In the first quarter of fiscal 2026, Applied Materials delivered strong performance, supported by robust global demand for advanced chip manufacturing equipment. Management expects semiconductor equipment spending to grow more than 20% this year, driven by leading-edge logic, high-bandwidth memory, and advanced packaging — all critical for AI workloads.

As chipmakers race to build more powerful processors and memory for generative AI, they must adopt complex technologies like Gate-All-Around transistors and advanced DRAM architectures. Applied sits at the center of this transition, providing the tools that enable these breakthroughs.

The company also announced the opening of its EPIC Center, designed to accelerate collaboration and next-generation technology development, including partnerships with major players like Samsung.

While export restrictions and capacity constraints remain risks, the broader trend is clear: AI is driving a structural upgrade cycle in semiconductor manufacturing.

The takeaway? Applied Materials isn’t just riding the AI wave — it’s supplying the tools that make the wave possible.

Today we’re looking at Equinix, one of the most critical — and often overlooked — players in the AI infrastructure buildout.

In 2025, Equinix delivered strong financial performance, with net income up significantly year over year. But the real story isn’t just earnings growth — it’s structural demand. As artificial intelligence scales, data centers must handle higher power density, advanced cooling, and ultra-low-latency interconnection.

Equinix operates more than just facilities. It runs a global ecosystem serving over 10,000 customers — from hyperscalers to enterprises — creating powerful network effects through interconnection. The more participants join, the more valuable the platform becomes.

However, AI infrastructure is capital intensive. As a REIT, Equinix must balance growth investments with disciplined capital allocation and dividend commitments. Rising interest rates and energy constraints remain key variables.

The takeaway is simple: AI may be digital, but its foundation is physical. And Equinix owns part of that foundation.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Today we’re talking about Arista Networks and how artificial intelligence is reshaping its growth story.

2025 was a record year for Arista. Revenue surged, driven primarily by explosive demand for AI networking infrastructure. The company generated roughly $9 billion in total revenue, with AI data center networking emerging as one of the strongest contributors. What’s most important is not just the growth itself — it’s the acceleration. Management raised its 2026 outlook, targeting around $11.25 billion in revenue, supported by unprecedented demand for high-performance Ethernet solutions.

AI is fundamentally changing the architecture of data centers. Training large language models and running inference at scale require ultra-low latency, massive bandwidth, and high reliability. Arista’s 800G Ethernet platforms are becoming critical infrastructure for hyperscalers building clusters with tens of thousands of GPUs. The transition toward 1.6T networking is already in development — a signal that this is not a short cycle, but a structural shift.

Another key theme is “Arista 2.0.” The company is moving beyond switching hardware into AI-driven network automation. Its AVA platform uses intelligent systems to automate troubleshooting, optimize performance, and manage increasingly complex AI environments. This software layer strengthens customer stickiness and expands margins over time.

However, growth at this scale comes with operational challenges. Supply chain constraints and semiconductor commitments remain significant. Arista has increased long-term purchase commitments to secure capacity, aiming to protect gross margins in the low-60% range despite component volatility.

Strategically, Arista is positioning itself as the Ethernet backbone of generative AI. Instead of proprietary interconnect standards, it is betting on open Ethernet ecosystems — collaborating across the industry to scale AI clusters efficiently.

The takeaway? Arista is no longer just a cloud networking company. It is becoming foundational infrastructure for the AI era. And if AI workloads continue scaling as expected, Arista stands at the center of one of the most powerful secular trends in technology.

This is not just cyclical growth. It’s architectural transformation.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Vertiv closed 2025 with exceptional momentum, driven by one of the most powerful infrastructure cycles in recent history: artificial intelligence.

Orders surged across the year, reflecting massive demand for AI-ready data centers. Revenue, operating income, and free cash flow all expanded significantly, supported by strong execution and disciplined cost management. Importantly, backlog growth signals sustained visibility into 2026.

Vertiv sits at a critical intersection of the AI buildout. As hyperscalers and enterprises deploy increasingly power-dense AI clusters, the need for advanced cooling systems, power management, and energy-efficient infrastructure becomes non-negotiable. AI workloads require far more electricity and thermal control than traditional compute — and Vertiv provides the backbone solutions.

Management expects continued growth into 2026, with global capacity expansion and operational scaling to meet structural demand.

The takeaway is clear: Vertiv is not a cyclical rebound story — it is leveraged to a structural AI infrastructure wave, where power, cooling, and reliability become strategic assets in the digital economy.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Cisco delivered a strong second quarter for fiscal 2026, with results coming in ahead of expectations and reinforcing its role in the AI infrastructure cycle.

Revenue growth accelerated, driven by robust demand for advanced networking, data center upgrades, and the company’s high-performance Silicon One chips. As enterprises and hyperscalers scale AI workloads, they are increasingly forced to modernize their network architecture — and Cisco is positioning itself at the center of that upgrade cycle.

Management highlighted the shift toward what it calls “Agentic AI,” where autonomous AI systems require secure, high-speed, low-latency connectivity. This evolution is driving demand not just for raw bandwidth, but for deeply integrated security and intelligent network management.

Strategic collaboration with NVIDIA further strengthens Cisco’s position in high-performance computing and AI clusters. At the same time, the company raised its full-year outlook, reflecting confidence in sustained AI-driven infrastructure spending.

The key takeaway is clear: Cisco is evolving from a traditional networking vendor into a foundational layer of AI-era digital infrastructure — combining hardware, software, and security into a more resilient, recurring revenue model.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Nebius Group closed 2025 with exponential revenue growth and its first adjusted EBITDA positive at the group level — a clear signal that scale is starting to translate into operating leverage.

The company’s strategy is bold: build one of the largest dedicated AI cloud platforms in the world. Nebius is aggressively expanding global data center capacity, targeting over 3 gigawatts of contracted power by the end of 2026, with a roadmap aimed at reaching $7 to $9 billion in annualized recurring revenue.

Unlike traditional cloud providers, Nebius is positioning itself as a pure-play AI infrastructure platform. It is deploying next-generation NVIDIA systems, building proprietary data centers, and enhancing its software layer through acquisitions like Tavily, strengthening its agentic AI and inference capabilities.

Massive capital expenditure — projected between $16 and $20 billion — underpins this expansion. Management believes demand for AI compute continues to exceed supply, allowing Nebius to operate near full capacity with long-term contracts and improving pricing stability.

The takeaway is clear: Nebius is making a high-conviction bet on AI infrastructure scarcity. Execution, capital discipline, and sustained demand will determine whether this aggressive scale-up converts into durable profitability.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.

Cloudflare closed 2025 with another year of strong execution, confirming its role as a critical layer of the modern internet.

Full-year revenues reached $2.17 billion, up roughly 30%, driven by accelerating demand from large enterprise customers. In Q4, growth was fueled by security services, edge computing, and especially Cloudflare Workers, which are increasingly used to deploy AI-driven applications closer to users.

Management highlighted the rise of what it calls the Agentic Internet — a world where autonomous AI agents interact, transact, and operate in real time. This shift significantly increases the need for low-latency compute, security, and global connectivity, all areas where Cloudflare is structurally positioned.

Importantly, growth is becoming higher quality. Large customer additions are increasing, contract sizes are expanding, and Cloudflare is embedding itself deeper into customers’ core architectures rather than acting as a peripheral service.

The takeaway is clear: Cloudflare is no longer just a content delivery or security provider. It is evolving into foundational infrastructure for the AI-native internet — a long-term trend where scale, performance, and trust truly matter.

Disclaimer

This content is provided for informational and educational purposes only.

It does not constitute financial, investment, legal, or tax advice.

Any opinions expressed are based on publicly available information and are not a recommendation to buy or sell any security.

Listeners are solely responsible for their own investment decisions.