Pinned Tweet
Most people are learning about memory stocks through AI news or their favorite financial influencers, who themselves only recently discovered the sector.
Here’s my edge: I’ve actually bought memory directly from these suppliers. During my time at a Fortune 50 tech company, I worked as a Memory Operational Commodity Manager, giving me firsthand knowledge of how the industry operates and how the major semiconductor suppliers think.
That direct relationship with their largest clients and understanding of their supply chain constraints, is something you can’t get from Twitter threads or financial media.
Recently, major tech firms are now offering to pay for SK Hynix’s equipment capex. If you understand semiconductor economics, you know this is absolutely unprecedented. It signals desperation for supply and is massively bullish for the memory supercycle ahead.
By subscribing, you’re getting analysis from someone who actually lived inside this industry not just someone reading about it. That’s the difference between surface-level commentary and real conviction. Join the LRMI community today! $MU $DRAM $NVDA $SMH $SNDK $WDC $SKHYNIX
lrmi.ai/p/we-dont-own-enough…
$ORCL is heading back to $200
Oracle $ORCL is projecting a nearly $54B turnaround in annual free cash flow over 3 years
Now expects to turn FCF positive in the H2 of FY28 & generate at least $30B in FY29
Expects its massive AI data center spending to peak in FY27-FY28 before declining in FY29
By FY29, Oracle says its growth will be self-funded.
There’s a lot of asymmetric information out there rn
You just have to find it
7 retweeted
Revenue per GPU can increase as token prices decline if per GPU token throughput increases enough.
So both Anthropic revenue *and* margins can go up as token prices go down *if* GPU prices go up less than the product of (token price decline * per GPU throughput increase).
But I have super enjoyed all the commentary to the contrary!