NVDA, MU, INTC & Other Names Ivan Feinseth Sees Leading Generational Tech Cycle

Micron posted revenue growth of more than 250% and earnings growth above 800%, driven by memory demand materially exceeding supply. Ivan Feinseth expects a prolonged AI investment cycle and identifies Nvidia, Intel, and SanDisk as core AI-infrastructure beneficiaries. He also characterizes MongoDB's selloff and Microsoft's recovery as buying opportunities.

Analysis

The key investable issue is not headline memory growth but whether HBM and server-DRAM mix can remain tight as capacity is redirected from commodity DRAM. MU is the cleanest operating-leverage expression: incremental HBM qualification and higher-value server mix can expand gross margin faster than revenue over the next 2-3 quarters. The offset is that memory equities typically discount the next supply response early; any indication that Samsung or SK Hynix has materially improved HBM yields or that hyperscaler inventory has normalized would compress MU's multiple before reported pricing weakens.

NVDA remains the demand-pull beneficiary, but its memory suppliers' strength is not automatically incremental upside for NVDA after an extended AI-capex cycle. The more differentiated second-order read-through is MSFT: sustained component tightness raises the cost of AI capacity and makes Copilot/agent monetization milestones more important to defend cloud margins over the next 6-18 months. INTC is a weaker AI-infrastructure expression than the group implies; absent externally validated foundry wins and improving process economics, elevated industry AI spending may instead widen its capital-intensity disadvantage versus TSMC and NVDA's ecosystem.

MDB's drawdown is only a buying opportunity if Atlas consumption reaccelerates independently of AI narrative enthusiasm. AI workloads can create database demand, but they can also shift spending toward hyperscaler-native data services and vector/database alternatives; watch net revenue retention, Atlas growth, and forward operating-margin guidance rather than treating a lower share price as evidence of value. SNDK offers more torque to a NAND recovery but also carries materially greater pricing and inventory-cycle risk than HBM-exposed MU.

Consensus is likely underpricing the duration of HBM constraints while overgeneralizing that constraint across all memory categories. Over the next 1-3 months, supplier commentary on HBM qualification, lead times, and capex is the catalyst; over 6-18 months, the decisive variable is whether AI demand converts from concentrated hyperscaler buildout into broad enterprise inference demand. A break in MU pricing expectations, reduced hyperscaler capex guidance, or a material HBM supply-yield improvement would falsify the bullish memory thesis.

Key Decisions for Investors

• Initiate a 3-6 month long MU / short SNDK pair: favor HBM and server-DRAM mix over commodity NAND beta. Size for a 10-15% adverse spread move; take profits if MU's next-quarter pricing or gross-margin outlook stops rising, regardless of reported results.
• Maintain NVDA as the preferred core AI exposure, but avoid adding solely on memory-strength headlines. Add on 8-12% market-driven pullbacks with a 6-12 month horizon; reduce if hyperscaler capex commentary turns from capacity constraint to utilization concern.
• Use MSFT as the lower-volatility AI monetization position rather than INTC: accumulate over 6-18 months only if Azure growth and commercial remaining-performance-obligation trends remain resilient while AI-related capex does not drive a sustained margin-guide reset.
• Do not buy MDB on the pullback without confirmation from the next earnings release: require stabilization or improvement in Atlas growth and net revenue retention. If either deteriorates, treat the stock as vulnerable to further multiple compression rather than an AI beneficiary.
• Keep INTC underweight versus NVDA/MSFT until foundry customer traction, process execution, and gross-margin recovery are independently evidenced. A credible external foundry contract plus improving margin guidance would be the trigger to revisit the shortfall thesis.