Semiconductor Engineer, Ph.D. 🇰🇷🇺🇸 📍 San Jose Startup & Big Tech Vet John's Papa Silicon Valley Trivopaedia: Tech, Money, Food & Chips. Life in the SV
San Jose, CA
Joined December 2009
- Tweets21.9K
- Following1.8K
- Followers15.5K
- Likes33.6K
Pinned Tweet
This morning I wrote why I stayed on X as a 30-year semiconductor engineer.
I do not float random tickers and numbers as I walk by. I invest where I work. I put real money only in the names I can defend with people still on the floor. Then I come back to those same names and share what I actually know.
That is the account.
Since I came back to X in January I have posted the same short list — tickers, company names, and Korean.
$NVDA
$DELL
Memory: $DRAM (listed Apr 2), SK hynix / $SKHY, $MU, SanDisk $SNDK
$CBRS (IPO May 14)
$AMD
$AVGO
Tape since Jan 1, except the two that listed this year (from listing day):
$NVDA +17% YTD
$DELL +350% YTD, about +11% today
$DRAM +117% since Apr 2
$MU +240% YTD
$AMD +140% YTD
$AVGO +5% YTD
$CBRS about flat from the $185 IPO, deep red from the day-one print
$SKHY ADR listed in July — my book is still green
$SNDK one of the strongest names in my own memory sleeve
$NVDA
Still the center of the board. Everything else I own is a satellite around this stack. My own cost basis is about 16x. Almost all of the position is gain, not principal. I do not need a taxable sale, and I have not found a better home for that capital. Custom silicon is cost control. It is not a full NVIDIA replacement.
$DELL
This is not a PC ticker. I used to own $SMCI as the one-top NVIDIA AI server name. Then the label-swap mess. I sit near people who have known that CEO and VP for decades. I heard the dawn FBI story as it landed. A stock can still run. I sold the whole position. Too many other choices to stay where I no longer trust the people.
I moved it into Dell. Pure U.S. name. Trump has said "go buy a Dell" more than once, and it shows up in his disclosed book. I am not here for XPS. I am here for Dell as an AI server company — $95B backlog, Vera Rubin racks with CoreWeave. Today is another +11% session. Thesis has not changed.
Memory
I stay long the bottleneck. $DRAM I bought on listing day, April 2. As a U.S. investor I did not have a clean single ticker for Samsung + SK hynix + Micron. The basket let me put real weight on it. Over +100% since listing. I still tell Korean investors not to copy that ticker. Buy the Seoul names at home and keep $MU / $SNDK for the rest.
Then IBKR omnibus for SK hynix ordinary shares, $SKHY ADR, $MU and SanDisk as singles. HBM yield, 1c mix, CPO, the new San Jose building — that is why SK hynix gets so many of my posts. Micron is the U.S. third leg. SanDisk fills the NAND hole the ETF underweights. Memory stays strong for a while. Nvidia's own call keeps repeating the same constraint.
$CBRS
This is the one that still hurts in the book. I was in on the idea before the IPO. The deal was too hot. I got a sliver, then bought more in the open market. That average is now my largest loss.
I still praise the architecture as an engineer. Harvest, not a perfect wafer. Defect-tolerant tiles. Complementary to GPUs, not a replacement. That is why I keep writing it next to NVIDIA and memory.
I visit Cerebras HQ next week, including the server room. I want to see the idea closer than a slide.
$AMD
Almost the only company that owns GPU, CPU, and ASIC in one house. Second in GPUs to NVIDIA. Second in CPUs to Intel. That full stack is the point. The Xilinx FPGA franchise is still the prototyping top for chips in front of those GPUs and ASICs.
$AVGO
Cable-modem days, then WiFi and Bluetooth. VMware made it a software company too. This cycle the payout is the ASIC / SoC design-partner seat — Jalapeño and the custom-silicon wave.
Everyday life stays. These names stay.
Not financial advice. I own what I write. I write what I work on.
I’ve spent nearly 30 years in the semiconductor industry as an engineer and engineering manager, mostly keeping my head down and just doing the work. That’s also why I ended up staying on X.
Back then, when someone asked “What do you actually do?”, it was hard to explain. In the last few years the money has poured into semiconductors and AI, and now even my 80-year-old parents know the semi company names and ask about them.
So my investing naturally stayed in the area I know best — the one where I can exchange real information and opinions with people still working in the industry.
Translating the semiconductor technology into language regular investors can use is still not easy.
I’ll post some everyday stuff too, but I’ll try to explain semiconductor technology and semiconductor investing as simply as I can.
NVIDIA quietly cut Vera Rubin's HBM4 bandwidth by about 13%.
Per-GPU HBM4 bandwidth on the datasheet went from 22 TB/s to 19.2 TB/s.
NVL72 went from 1,580 TB/s to 1,400 TB/s.
Capacity stays at 288GB, and NVFP4 compute is unchanged.
Same 2048-bit interface, lower pin rate.
19.2 TB/s works out to roughly 9.4 Gbps per pin.
On HBM4, the last Gbps is where base die, TSV channel, and interposer margin all get expensive.
Relaxing the pin target should widen the pool of stacks that pass qual and lift effective yield.
The trade-off is that the fastest-binning supplier loses some of its speed premium, and a chaser like Micron gets an easier bar.
The yield and qual read is industry interpretation, not an NVIDIA statement.
No supplier yield numbers were disclosed.
$NVDA $MU 000660 / $SKHY
$TSLA finally back above $400. First time since mid-July. +28% over the past 3 months.
Thanks!!
I connect Grok Bot to my account!
Last week, we removed the need for a developer account and X API credits to use the X connector in Grok Bot
Ask @bot to manage your bookmarks, go over your DMs, or draft an article for you free of additional charge
Papa Johns retweeted
꼰대의 딴지일지 모르겠지만, 카카오페이에 계정을 못 만드는 사람이라 실제 이 화면을 본 적이 없었는데, 이건 카카오가 한국 표준어를 혼란스럽게 만들고 있는 것 아닌가요?
아래 캡쳐에 있는 "송금보내요", "송금 받기" 이건 모두 어색한데요.
‘송금(送金)’은 한자로 ‘보낼 송 + 돈 금’이라 이미 ‘돈을 보내다’는 뜻입니다.
"돈 보내요 보내요"가 되어버립니다.
"송금 받기"는 "돈 보내기 받기"라는 말도 안 되는 단어가 되어 버립니다.
카카오 같은 대형 회사가 이런 걸 모르고 쓴 것일까요?
이런 것을 보면 어색한 것은 저만 그런 것일까요?
Apple confirms a small (?) number of iPhone Duos have inner display issues (often cracks/lines from the hinge) and is offering replacements via AppleCare.
theverge.com/tech/1018760/ip…
Amkor $AMKR reports tonight after the close.
Guide: $1.95–2.05B sales, 18.5–19.5% gross margin, $0.72–0.82 EPS. Options imply ±13.3%.
Amkor has a strong position in AI-era advanced packaging.
The name itself is America (AM) plus Korea (KOR).
Packaging is where the AI buildout becomes physical — capacity, not announcements, is the number to watch.
Cadence $CDNS reports tonight after the close.
Consensus about $2.05 EPS on about $1.61B revenue, up about 20% YoY. Guide: $1.595-1.625B. Options imply ±7.5%.
EDA from Cadence is the everyday design tool for me, and the quiet read before Wednesday’s hyperscaler capex prints — if design tools slow, the buildout story gets a check.
Papa Johns retweeted
대입시험을 위해서 학교에서 한자를 배우던 세대 사람이라 한자를 사용하는 한국어의 이상한 표현이 눈에 거슬리게 들어오는 사람이라 그런 것 같네요.
카카오페이가 생긴 지도 몇 년 되었을 텐데, 이미 저게 표준어인지 잘못 알고 쓰는 사람이 카카오 덕분(?)에 많아졌겠네요
꼰대의 딴지일지 모르겠지만, 카카오페이에 계정을 못 만드는 사람이라 실제 이 화면을 본 적이 없었는데, 이건 카카오가 한국 표준어를 혼란스럽게 만들고 있는 것 아닌가요?
아래 캡쳐에 있는 "송금보내요", "송금 받기" 이건 모두 어색한데요.
‘송금(送金)’은 한자로 ‘보낼 송 + 돈 금’이라 이미 ‘돈을 보내다’는 뜻입니다.
"돈 보내요 보내요"가 되어버립니다.
"송금 받기"는 "돈 보내기 받기"라는 말도 안 되는 단어가 되어 버립니다.
카카오 같은 대형 회사가 이런 걸 모르고 쓴 것일까요?
이런 것을 보면 어색한 것은 저만 그런 것일까요?
Home owner + Referee + Player.
The triple combo that kills fair financial discourse.
I've always defended X as the only platform to be on for financial discourse.
But it's getting harder to when the owner constantly amplifies narratives of collapse… of companies whose interests conflict with his own at $SPCX or $TSLA.
Would people be comfortable if $NVDA Jensen Huang owned $META?
And then the timelines were dominated by his amplified narratives of $AVGO failing?
I can't see how that would be healthy for grassroots discussion.
July 12, 2017 email from Ilya Sutskever to Elon Musk @elonmusk & Greg Brockman @gdb
(“The business of building AGI”)
At the time, Jim Keller @jimkxa (then VP of Autopilot Hardware Engineering at $TSLA ; later Intel SVP, now Tenstorrent CEO) said a 4-chip card equivalent to TPU 3.0 would take him about 2 years.
Sutskever’s take:
> “The Cerebras @cerebras design is far ahead of both of these, and if they’re real then having exclusive access to them would put us far ahead of the competition.” - (pink highlight on the attachment)
Source: storage.courtlistener.com/re…
(Exhibit in Musk v. Altman)
$CBRS
In a July 18, 2017 email, Shivon Zilis briefed Elon Musk @elonmusk and Sam Teller on discussions with Greg Brockman and Ilya Sutskever.
They viewed Cerebras @cerebras as a high-beta experiment: if it worked, it would be game-changing, and if Google got it, the AGI race would be effectively over. Even starting a billion-core system that day would leave them 18 months behind.
The email also noted that Andrew Feldman @andrewdfeldman had reached out and a meeting was scheduled for the following week.
This email was released as evidence in the Musk v. Altman lawsuit (Exhibit 58).
Almost 10 years ago, Elon and OpenAI were already seriously evaluating Cerebras $CBRS 🫣
👇 Source: storage.courtlistener.com/re…
In a July 18, 2017 email, Shivon Zilis briefed Elon Musk @elonmusk and Sam Teller on discussions with Greg Brockman and Ilya Sutskever.
They viewed Cerebras @cerebras as a high-beta experiment: if it worked, it would be game-changing, and if Google got it, the AGI race would be effectively over. Even starting a billion-core system that day would leave them 18 months behind.
The email also noted that Andrew Feldman @andrewdfeldman had reached out and a meeting was scheduled for the following week.
This email was released as evidence in the Musk v. Altman lawsuit (Exhibit 58).
Almost 10 years ago, Elon and OpenAI were already seriously evaluating Cerebras $CBRS 🫣
👇 Source: storage.courtlistener.com/re…
Papa Johns retweeted
Most takes on the Solidigm IPO halt so far are from the SK hynix investor side.
Talking with engineers around San Jose, I hear a different story.
For people who joined Solidigm with an IPO exit in mind, this is a heavy letdown.
No decision can satisfy everyone, but that disappointment is real.
A halted IPO can be a win for the parent's shareholders and a retention problem for the subsidiary at the same time.
In NAND and enterprise SSD, experienced controller and firmware engineers are hard to replace.
What SK hynix offers those people next matters as much as the IPO decision itself.
SK hynix has decided to withdraw the Nasdaq IPO plan for its US NAND subsidiary Solidigm, according to an exclusive report from Hankyung.
The decision comes after strong shareholder pushback over dual-listing concerns, potential dilution of enterprise value, and reduced ownership for existing shareholders. The related pre-IPO fundraising process has also been halted.
An official announcement is expected soon.
hankyung.com/article/2026102…
Papa Johns retweeted
Papa John's portfolio update as of Oct 25 (after Friday close)
Past ~1 month: investment assets roughly -5%.
Latest changes: significant rotation into $CBRS from memory investments.
- Roth IRA: fully exited $DRAM with meaningful gain (tax-free account, so no short-term gain tax concern even under 1 year) and moved the proceeds into $CBRS.
- Robinhood: sold the $SKHY ADR while it was still green (small position). Last Friday rotated that into more $CBRS.
- Fidelity BrokerageLink: cleared $MU and other names, consolidated fully into $CBRS.
In effect, I leaned harder into $CBRS on the recent decline — a stronger upside bet.
Current top allocation (investments):
$NVDA 24.0% · Chase JPM 15.1% · $CBRS 12.5% · $AAPL 6.7% · $TSLA 6.0% · $ZM 4.8% · $AVGO 3.4% · $PLTR 3.1%
Papa Johns' Portfolio review
It's been about 3 months since I posted this.
Time for a quick review and recall.
Here’s my current portfolio showing the investment performance over the past 3 months.
The stock market has been quite dynamic during this period.
Especially the Korean market — memory-related stocks seemed to lead a big part of the overall market flow.
From Korean news, it looks like FOMO kicked in, bringing in a lot of new investors, only for many of them to take hard-to-recover losses in last month’s sharp drop and leave.
The original post was my answer, as one of the working semiconductor PhDs, to this question:
“I have a pure curiosity question… Do current PhDs actually invest a lot in semiconductors? And do you get good returns based on that specialized knowledge?”
Looking back at the review, I didn’t make a huge profit over these 3 months.
But during a period with big swings, overall investment assets increased slightly with no losses.
---
Comparing then and now from 3 months ago.
I’ve spent my entire career in semiconductor design. Just as I start thinking about retirement, semiconductors have become extremely hot.
The majority of my investment performance still comes from semiconductor-related stocks.
- $NVDA +1,607.54% (still the overwhelming top performer)
- $AAPL +180.01%
- $AMD +130.92%
- $SNDK +124.20% (entered 3 months ago and rode the memory bottleneck theme strongly)
- $TSLA +115.61%
- $DRAM +107.14% (bought on the launch day out of conviction in memory. At the time it was a convenient way for U.S. investors to get exposure to Samsung and SK Hynix)
- $PLTR +98.57%
- $AVGO +95.89%
- $DELL +89.82% (as I’ve mentioned several times in my posts, I viewed it as the optimal play for AI GPU servers. After the so-called “label swap” incident with $SMCI during the last GTC period, I sold all my $SMCI and moved into DELL as an alternative AI server supplier. It’s been a quite satisfactory investment)
- $AMZN +64.15%
- $MSFT +60.25%
- $SKHY +24.79%
- $ZM -1.28%
- $GOOGL -4.92%
- $CBRS -15.50%
- $SPCX -16.17%
I participated in the IPOs of both $CBRS and $SPCX.
Those IPO shares are left untouched in a separate account (as the 2nd image).
Fortunately they are still up +24.60% and +4.92% respectively.
The additional shares I bought later in my regular trading account are unfortunately in the red right now.
But I’m holding them with the hope that their future impact will be significant.
Of course, besides the accounts I captured and introduced here, I have additional accounts opened and run for separate purposes.
For example, using an $IBKR account for direct trading of Korean stocks, or managing part of my 401(k) with direct stock picks, and so on…
---
As I said 3 months ago,
knowing a lot of semiconductor companies that others don’t know can itself be a powerful investment weapon.
But if you only look at the technology and miss external variables, you can also get badly burned.
Looking back, the stocks that lost me significant money were mostly ones I didn’t really understand and just bought because “people said it was good.”
The same held true over these past 3 months.
Many people jumped in with FOMO during the memory rally and got hurt in the sharp drop,
but I only moved within the trends I already understood (memory bottleneck, AI GPU dominance), so I was able to get through without major losses.
Please don’t ask me “Should I buy? Sell? How high do you think it will go?”
I don’t know either 🤪
I’ll keep sharing semiconductor news, company updates, and technical stories that are useful for investing whenever something interesting catches my eye.
If you follow me, you’ll continue to see real-time answers to the question
“Do working semiconductor PhDs actually invest a lot in the sector?”
Papa Johns retweeted
Papa John's portfolio buckets as of Oct 25 (after Friday close)
~80% taxable accounts
~20% tax-advantaged (401k, Roth IRA, 529)
Most of it is direct stock picking. As a small buffer to my own investing, I also keep some in Chase JPM funds managed by professionals.
Account mix:
Robinhood 62.0% · Chase 14.6% · Fidelity 12.3% · Vanguard 529 6.5% · IBKR/SoFi/Cash 4.5%
Fidelity includes BrokerageLink + 401(k). Robinhood includes a Roth IRA.
Papa John's portfolio update as of Oct 25 (after Friday close)
Past ~1 month: investment assets roughly -5%.
Latest changes: significant rotation into $CBRS from memory investments.
- Roth IRA: fully exited $DRAM with meaningful gain (tax-free account, so no short-term gain tax concern even under 1 year) and moved the proceeds into $CBRS.
- Robinhood: sold the $SKHY ADR while it was still green (small position). Last Friday rotated that into more $CBRS.
- Fidelity BrokerageLink: cleared $MU and other names, consolidated fully into $CBRS.
In effect, I leaned harder into $CBRS on the recent decline — a stronger upside bet.
Current top allocation (investments):
$NVDA 24.0% · Chase JPM 15.1% · $CBRS 12.5% · $AAPL 6.7% · $TSLA 6.0% · $ZM 4.8% · $AVGO 3.4% · $PLTR 3.1%
These are what @Grok knows about me.
