←
Photon Capital20,139 posts
Photon Capital avatar

Photon Capital

@PhotonCap

Seeing tech through a different wavelength. Photonics & semiconductor research photoncap-map.pages.dev/

◷ Joined April 2025▧ 2,580 Photos and videos
Photon Capital avatar
Photon Capital@PhotonCap·9h

At TomorrowX, I listened to Lisa Su of AMD $AMD, Michael Dell of Dell $DELL, David Solomon of Goldman Sachs $GS, Jitendra Mohan of Astera Labs $ALAB and others. I didn’t catch every session, but the conversations I heard were largely focused on the long-term promise of technology. I share that optimism. The computing ecosystem Jensen Huang and NVIDIA $NVDA helped build has made today’s AI and infrastructure possible. I believe it can also help improve lives through drug discovery, cancer treatment and more. The potential downsides of AI were even addressed in Korea’s recent college entrance exam. Those concerns are real. But I hope we can move beyond fear and find ways for technology and policy to work together so that more people can benefit. The TomorrowX opening video has stayed with me. It traced humanity’s journey from the rise of civilization and the invention of the wheel to the steam engine, semiconductors, GPUs and AI. People have surely feared new technologies before. Yet we have also used them to build better lives. AI is another tool in that journey. What matters is how we use it to benefit humanity and help people live better, more peaceful lives. Investing, however, is a different question. Stanley Druckenmiller said he wouldn’t dare short equities with so much innovation underway. But he also couldn’t remember a period when he had less clarity about where things would stand in 12 months. For now, he’s focused on preserving capital and mostly staying away from equities. “I play big when I’m hot, when I’m cold I do little bunts.” My own gains have fallen significantly from their peak. I’m still excited about the future of technology, but honestly, I’m not sure what to do with my portfolio right now. When even Druckenmiller is sitting much of it out, you realize how difficult this market really is.

STANLEY DRUCKENMILLER ON THE MARKET: "I can't remember a period where I had less clarity on where we're gonna be in 12 months." "We're in capital preservation, trying to not lose money as opposed to make money." "We're not long equities & we're not short." "I'm really alert to the fact that the second half of '27 could be economically disappointing." "I wouldn't dare go short with all the innovation going on." "Those stocks are oversold, they're washed out. We have some small positions but we're basically staying away." We are short Copper, “The Fed is not cutting, they’re hiking.” "I'm not playing. I play big when I'm hot, when I'm cold I do little bunts."

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·10h

$DECK

The GOAT, @standuquesne, at @TomorrowXSummit: • Thanked @GavinSBaker for identifying the AI opportunity early. • Nine months ago, he was long copper and equities, short bonds, and thought healthcare would break out. Today, he owns almost none of those positions. • At the time, he had conviction: the Fed funds rate was around 3.5%, monetary policy was loose, and commodities were hitting new highs. A recession within 6–12 months seemed unlikely. • Claude Code had people excited about accelerating AI progress, adding to his bullish view on equities and the economy. • Then war started. AI capex remains strong, but much of the broader picture has flipped. • He sees a change in the Fed regime, with the Fed hiking and Warsh unhappy with what has been happening. • Commodities have broken out, but they are now short copper. Credit spreads are widening. • They are virtually flat equities: not long, but reluctant to be short with some things oversold and so much innovation underway. • After 50 years of investing, he has less clarity on where we’ll be in 12 months than ever before. • They have some small positions in the dollar and are largely staying on the sidelines. • Bonds and currencies tend to make their biggest moves when equities are falling. • They are in capital preservation mode. The priority is to avoid losing money. • After a great three or four year run, they’re taking a break. It could last two days or six months. • Their risk management is unusual: they don’t use VaR. In periods of chaos, the usual correlations across asset classes can break down. • He believes in streaks. A big part of the job is recognising when you’re seeing the ball clearly and when you’re not. • Given how commodities are behaving, he thinks the second half of 2027 could be disappointing. • He’s looking for AI use cases and doesn’t think infrastructure will be the main engine forever. AI labs could be among the best investments if they begin selling completed tasks rather than tokens. • He worries about the deterioration of US K–12 education. The US has worked around this by importing talent, but immigration policy could make that harder. • He also worries about US fiscal challenges and sees the possibility of a crisis forcing major fiscal reform within the next decade. • Investing edge isn’t simply about IQ. Intellectual humility matters: smart people can become attached to a thesis and be too slow to adjust. • Put your eggs in one basket and watch the basket very carefully. • What separates good investors from great ones? Curiosity, passion, discipline, an extreme competitive drive and, often, little work-life balance. • Every buyer has a seller. Understand who is on the other side of your trade. • He’ll make a meaningful initial bet with perhaps 15% of the information others might want, then increase the position as his conviction grows. • His longer-term predictions: a 20% chance of fusion energy, a 70% chance of data centres in space, and cancer becoming a managed disease within ten years.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·15h

- $NVDA announced an amendment to its existing substrate supply agreement with Samsung Electro-Mechanics ($009150) on November 17. - This expands their long-term partnership announced on October 21, rather than establishing a new supplier relationship. - Samsung Electro-Mechanics supplies ABF package substrates for AI processors and high-speed networking from its Sejong and Busan facilities. - $NVDA also signed a new agreement with Kinsus to expand dedicated ABF substrate capacity by more than 3.5x. > NVIDIA is securing long-term advanced packaging and substrate capacity to address GPU production bottlenecks. > Samsung Electro-Mechanics is strengthening its position as a critical substrate partner across NVIDIA's AI computing and networking infrastructure. > This is an ABF substrate agreement, not an HBM supply deal. Samsung Electronics ($005930) and Samsung Electro-Mechanics ($009150) should be distinguished. > The key metrics are NVIDIA-dedicated capacity expansion and actual shipment growth.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·16h

- $NVDA returned $28.1 billion to shareholders in Q3 FY2027 through repurchases and dividends. - Board approved $150 billion share repurchase authorization in September 2026, with $227.3 billion remaining. - Vera Rubin platform described as off to explosive start; roadmap on track under interim CEO Colette Kress. > $NVDA's substantial shareholder return and aggressive repurchase authorization signal confidence in cash generation amid AI infrastructure demand. > The Vera Rubin CPU platform's early momentum suggests expansion beyond GPU-centric offerings.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·16h

- $NVDA expanded capacity agreements with six supply chain partners across substrates, advanced packaging, wafer, and contract manufacturing to support AI infrastructure demand. - Kinsus to expand advanced substrate capacity by 3.5x with production starting H2 2027; Samsung Electro-Mechanics amended existing substrate agreement for improved operations in South Korea. - $NVDA is funding multibillion-dollar advanced packaging expansions with SPIL in Taiwan and nearly $3 billion support for Amkor in Arizona and South Korea through multiyear agreements. > $NVDA is locking in long-term supply commitments across multiple vendors to secure substrate and packaging capacity for sustained AI accelerator production, reducing manufacturing bottlenecks that have historically constrained AI infrastructure deployment.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·16h

완전 공감 어느순간부터 효율성 따지며 효율왕이 되어감

유행도 숙제가 됐다. 김난도 교수가 얘기하는 1년 만에 싹 바뀐 한국인 소비 트렌드 총정리 1. 술값부터 끊음. 한국에서 술은 사람 사귀는 데 쓰는 돈이었는데, 이제는 무알콜 술이 트렌드 코리아 2027 10대 상품에 들어감. 2. 친구도 ROI로 따짐. "이 친구는 ROI 안 나온다"며 하나둘 손절하다 보니, 생일에 좋아요는 많은데 촛불 불어 줄 사람은 적음. 3. 아낀 돈은 나한테 씀. 예전엔 주름이 생겨야 피부 관리를 했는데, 이제는 20대가 적금 들듯 피부과 다니면서 50대 피부를 미리 챙김. 4. 헬스장은 이제 근육을 팖. 비만약이 살을 쉽게 빼 주니 손님이 줄었는데, 약을 먹으면 근육도 같이 빠져서 근육 만들기로 방향을 틂. 5. 5060도 자기한테 쓰기 시작함. 부모 모시고 자식 키우느라 못 쓰던 돈을 옷·피부·취미에 써서, 영수증만 보고 나이 맞히기가 어려워짐. 6. 동창회 대신 러닝크루에 나감. 결혼하고 애 낳는 시기가 다 달라져 동창끼리는 옛날 얘기밖에 할 게 없고, 새 사람은 취향 모임에서 만남. 7. 여행은 엄마랑 감. 손절하다 보면 결국 남는 건 가족이라, 엄마랑 아울렛 가고 할머니랑 영화 보는 2030이 늘었다고 함. 물론 다 좋아진 건 아님. 즐길 건 넘치는데 알고리즘은 비슷한 것만 보여 주고 다들 혼자 놀아서, 2027 첫 키워드가 '권태 사회'임. 친구 손절하고 마지막에 남은 게 엄마랑 아울렛이라니, 웃긴데 좀 짠함.

Photon Capital avatar
Photon Capital@PhotonCap·18h

Finally got it off but looking inside the plastic fastener, I can see some vicious-looking teeth. Good luck to everyone else trying to take off their bracelets tonight! @TomorrowXSummit

Post attachmentPost attachment

I’m being serious here. I’ve been wrestling with this for over 10 minutes and just can’t get it open. I took a look inside and there’s this sharp, pointed edge facing one way there’s literally no way to get past it, lol.

Photon Capital avatar
Photon Capital@PhotonCap·18h

중경외시였는데, 시가 외보다 앞이네요, 근데 포스텍은 어디있나요?

교육부 오피셜 건동홍부터는 지잡이다 눈물의 비명문대 보통 저런거 서열얘기나올까봐 가나다 순으로하는데 ㅋㅋ 교육부가 오피셜 박으면 어카냐 ㅋㅋㅋ 정부 오피셜 : 연대 > 고대😳

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·18h

I’m being serious here. I’ve been wrestling with this for over 10 minutes and just can’t get it open. I took a look inside and there’s this sharp, pointed edge facing one way there’s literally no way to get past it, lol.

how to release it? @TomorrowXSummit 😅

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·20h

- China controls roughly two-thirds of the global optical transceiver market, with Innolight and Eoptolink supplying major US hyperscalers. - $AMZN, $GOOGL, $META, and $MSFT warned that replacing Chinese suppliers could take up to a decade.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·20h

Astera Labs $ALAB CEO Jitendra Mohan at TomorrowX Summit. The reason he started the company nine years ago was fascinating. He anticipated that the world would need far more GPUs, and connecting them would become a major challenge. Astera started with retimers. Now it's expanding into fabric switches, memory connectivity, and optical interconnects. His comments on optics were particularly interesting. As GPU clusters grow and data rates increase, copper eventually runs into physical limitations. Optics becomes necessary. Astera is already expanding into silicon photonics, VCSEL-based connectivity, NPO, and CPO. I don't think $ALAB should be viewed as just a retimer company anymore. The AI infrastructure race is increasingly about how efficiently GPUs communicate with each other. More GPUs. More connectivity. More opportunity.

Post attachment

Next at TomorrowX Summit: Dina Powell McCormick, President & Vice Chairman of Meta $META. The conversation began with AI infrastructure. Data center development has become a political issue in the US, including during the recent midterm elections. Communities are concerned about electricity costs, jobs, and whether they will actually benefit from these massive investments. Dina highlighted Meta's investment in Richland Parish, Louisiana. Increased local tax revenues have helped fund substantial teacher bonuses, creating tangible benefits for the community. She also emphasized the importance of working with Louisiana Governor Jeff Landry. Data centers aren't just buildings filled with GPUs. They require electricians, fiber technicians, construction workers, and many other skilled professionals. AI infrastructure can create American jobs. But her bigger point was national security. If America cannot maintain its own computing infrastructure, its economic competitiveness and security could be at risk. Sovereign compute matters. She also expressed confidence in Meta's approach to AI safety and emphasized the importance of security. But the most powerful moment for me was a video about Meta AI glasses. A military veteran who lost his eyesight in a bunker explosion was shown using AI glasses to read letters, navigate restaurant menus with his wife, and find clothing. This is what meaningful AI looks like. Dina also spoke about raising six daughters in a world changing incredibly fast. Young people need to become AI-literate, but human skills, particularly emotional intelligence, may become even more valuable. As AI makes technical capabilities more accessible, uniquely human qualities could matter more. After listening to several CEOs today, I'm noticing a common theme. When discussing the future of AI, the conversation eventually returns to biology, medicine, science, jobs, and human lives. The real value of AI won't be measured solely by how many GPUs we deploy. It will be measured by what we accomplish with that compute.

Post attachment
Photon Capital avatar
Photon Capital@PhotonCap·20h

Next at TomorrowX Summit: Dina Powell McCormick, President & Vice Chairman of Meta $META. The conversation began with AI infrastructure. Data center development has become a political issue in the US, including during the recent midterm elections. Communities are concerned about electricity costs, jobs, and whether they will actually benefit from these massive investments. Dina highlighted Meta's investment in Richland Parish, Louisiana. Increased local tax revenues have helped fund substantial teacher bonuses, creating tangible benefits for the community. She also emphasized the importance of working with Louisiana Governor Jeff Landry. Data centers aren't just buildings filled with GPUs. They require electricians, fiber technicians, construction workers, and many other skilled professionals. AI infrastructure can create American jobs. But her bigger point was national security. If America cannot maintain its own computing infrastructure, its economic competitiveness and security could be at risk. Sovereign compute matters. She also expressed confidence in Meta's approach to AI safety and emphasized the importance of security. But the most powerful moment for me was a video about Meta AI glasses. A military veteran who lost his eyesight in a bunker explosion was shown using AI glasses to read letters, navigate restaurant menus with his wife, and find clothing. This is what meaningful AI looks like. Dina also spoke about raising six daughters in a world changing incredibly fast. Young people need to become AI-literate, but human skills, particularly emotional intelligence, may become even more valuable. As AI makes technical capabilities more accessible, uniquely human qualities could matter more. After listening to several CEOs today, I'm noticing a common theme. When discussing the future of AI, the conversation eventually returns to biology, medicine, science, jobs, and human lives. The real value of AI won't be measured solely by how many GPUs we deploy. It will be measured by what we accomplish with that compute.

Post attachment

Michael Dell, founder and CEO of Dell Technologies $DELL, at TomorrowX Summit. Gavin Baker of Atreides Management opened with an interesting question. Over the past five years, Dell has outperformed every Magnificent Seven stock except NVIDIA. Why haven't other server competitors achieved similar results? Michael Dell pointed to the company's longstanding foundation in servers, storage, and data center infrastructure. Long before the AI boom, Dell had already established deep relationships with national laboratories and enterprise customers. That foundation is now a major advantage in the AI infrastructure race. Then Gavin asked an even more interesting question. "What if we're at peak AI right now?" Michael Dell strongly disagreed. For that to happen, the fundamental thesis behind AI would have to be wrong. His reasoning was fascinating. Humans are naturally curious. We constantly want to discover more and solve problems we haven't yet figured out. Biology, materials science, and scientific research still have countless unanswered questions. And AI could drive much of the progress in these fields. He expects AI to become increasingly efficient while delivering substantial benefits to the US economy. What I find particularly interesting is that Dell doesn't build frontier AI models. Regardless of who wins the AI model race, computing infrastructure remains essential. While Wall Street continues debating peak AI capex, the founder of one of the world's largest server suppliers seems convinced that AI is still in its early stages. Perhaps the more important question isn't when AI infrastructure investment peaks, but how much scientific progress AI has yet to unlock.

Post attachmentPost attachment