Pre-Market
Pre-Market Briefing — 2026-05-25
May 25, 2026Pre-Market Briefing — 2026-05-25
Market Regime
QQQ $714.51 — +11.7% vs SMA50 ($639.70) · Risk On ✅
Portfolio
| Strategy | Value | Return | DD | Positions | Cash |
|---|---|---|---|---|---|
| 4-Pool | $1,947,942 | +1847.9% | -2.0% | 20 | 0% |
| V5.3 | $1,182,333 | +1082.3% | -2.3% | 18 | 0% |
Breakout Signals
No breakout signals today.
_Section section_options_radar unavailable.
Open Positions
| Ticker | Entry | Current | Gain% | Strategy |
|---|---|---|---|---|
| 000660.KS | $611.02 | $1416.06 | +131.8% | 4-Pool |
| 0981.HK | $8.21 | $9.54 | +16.2% | 4-Pool |
| 1347.HK | $12.06 | $16.45 | +36.4% | 4-Pool |
| 2513.HK | $102.10 | $129.82 | +27.2% | 4-Pool |
| APLD | $32.19 | $48.02 | +49.2% | 4-Pool |
| APP | $433.51 | $485.89 | +12.1% | 4-Pool |
| AVGO | $402.17 | $414.57 | +3.1% | 4-Pool |
| CIEN | $67.36 | $587.23 | +771.8% | 4-Pool |
| CIFR | $15.42 | $21.52 | +39.6% | 4-Pool |
| GFS | $47.80 | $81.35 | +70.2% | 4-Pool |
| LITE | $62.86 | $964.50 | +1434.4% | 4-Pool |
| LRCX | $246.49 | $302.24 | +22.6% | 4-Pool |
| MDB | $264.69 | $317.50 | +20.0% | 4-Pool |
| MP | $57.36 | $61.74 | +7.6% | 4-Pool |
| MU | $465.66 | $762.10 | +63.7% | 4-Pool |
| NOK | $11.30 | $14.18 | +25.5% | 4-Pool |
| PWR | $317.65 | $716.91 | +125.7% | 4-Pool |
| TER | $107.65 | $353.44 | +228.3% | 4-Pool |
| TSM | $365.90 | $407.15 | +11.3% | 4-Pool |
| WULF | $18.05 | $22.92 | +27.0% | 4-Pool |
| 000660.KS | $611.02 | $1416.06 | +131.8% | V5.3 |
| 005930.KS | $138.98 | $218.61 | +57.3% | V5.3 |
| 0981.HK | $8.21 | $9.54 | +16.2% | V5.3 |
| 9888.HK | $15.68 | $16.26 | +3.7% | V5.3 |
| ASML | $1421.05 | $1592.00 | +12.0% | V5.3 |
| CIEN | $67.36 | $587.23 | +771.8% | V5.3 |
| CIFR | $15.42 | $21.52 | +39.6% | V5.3 |
| CRDO | $119.59 | $193.39 | +61.7% | V5.3 |
| ETN | $400.44 | $381.51 | -4.7% | V5.3 |
| FCX | $49.15 | $62.31 | +26.8% | V5.3 |
| GFS | $47.80 | $81.35 | +70.2% | V5.3 |
| GOOGL | $317.24 | $387.66 | +22.2% | V5.3 |
| LRCX | $246.49 | $302.24 | +22.6% | V5.3 |
| MDB | $264.69 | $317.50 | +20.0% | V5.3 |
| PWR | $317.65 | $716.91 | +125.7% | V5.3 |
| QCOM | $156.00 | $213.41 | +36.8% | V5.3 |
| SHLS | $9.28 | $9.55 | +2.9% | V5.3 |
| TSM | $365.90 | $407.15 | +11.3% | V5.3 |
X Alpha — Analyst Chatter
Convergence Alerts — X + Breakout Scanner
| Ticker | Direction | Source | Breakout | Score | Thesis |
|---|---|---|---|---|---|
| TE | Long | @thevalueist | Base Recovery | 95 | 10x-long (3 mentions) in power semi/800VDC AI data-center th… |
| LPTH | Long | @fundmyfund | Base Recovery | 40 | Domain-aligned reaction to Lockheed Martin’s new NGSRI websi… |
| ## 1. MACRO & THEMATIC OVERVIEW |
Two independent analysts converge on the same multi-year AI infrastructure buildout. @aleabitoreddit highlights Jensen Huang’s projection of $3–4T annual AI capex by 2030 and flags upstream material and photonic bottlenecks. @jukan05 independently points to AI-server-driven demand spikes in passives (MLCC lead times) and a packaging shift toward embedded substrates and hybrid bonding that benefits Taiwanese suppliers. Both threads also surface geopolitical supply-chain bifurcation: Huawei’s new hybrid-bonding pathway for Kirin chips (fall launch) that bypasses leading-edge EUV, versus continued reliance on InP/SOI/TFLN and laser components sourced from a narrow set of non-Chinese specialists. Institutional passive flows into small-cap optical names are cited as an early confirmation signal.
## 2. HIGH-CONVICTION IDEAS
$AXTI — InP substrate choke-point for 800 VDC + CPO ramp
- Who: @aleabitoreddit (materials supply-chain mapping)
- Thesis: AXTI controls ~40 % of the indium/gallium/precursor/pBN crucible steps required for finished InP substrates; NVDA’s shift to 800 VDC and photonics buildout directly lifts substrate demand.
- Key data: 40 % of InP supply chain; $3–4 T annual AI capex by 2030.
- Valuation: Not provided.
- Catalyst: Photonics/CPO volume ramp 2026–2027.
- Engagement: [2185 L 253 RT 29 QT]
- Contrarian?: Against finished-substrate producers (Sumitomo); upstream leverage instead.
- Cross-references: Same analyst flags AXTI alongside SIVE/FOCI/MSSCorp in CPO chain.
$SIVE — Laser IP + index-inclusion catalyst
- Who: @aleabitoreddit (multiple posts)
- Thesis: Sivers supplies CW DFB lasers to AEVA (FMCW) and multiple CPO players; first-time inclusion in OMX Stockholm + pending MSCI/NASDAQ listings will drive $60 M+ passive inflows into existing float.
- Key data: $60 M+ Vanguard/BlackRock buying pressure next week; MSCI follow-on; NASDAQ listing imminent.
- Valuation: Not provided.
- Catalyst: Index inclusions (next week) then NASDAQ listing.
- Engagement: [728 L 68 RT 14 QT], [730 L 42 RT 6 QT], [420 L 6 RT 3 QT]
- Contrarian?: Early institutional ownership while still retail-dominated.
- Cross-references: Same analyst maps SIVE → AEVA → LG Innotek → Boston Dynamics and to AAPL/NOK/RTX defense.
$SOI / 6830 MSSCorp / 3363 FOCI — SOI, yield control, optical packaging
- Who: @aleabitoreddit
- Thesis: SOI dominates SOI substrates; MSSCorp holds deep CPO yield patents; FOCI benefits from TSM COUPE optical module ramp.
- Key data: “Substantive monopoly” positioning cited; TSM-linked institutional optimism.
- Valuation: Not provided.
- Catalyst: 2026–2027 CPO production ramp.
- Engagement: [1002 L 123 RT 10 QT], [369 L 17 RT 6 QT]
- Contrarian?: No direct disagreement in batch.
## 3. SUPPLY CHAIN & SECTOR MAP
- Lasers → CPO: $SIVE supplies DFB lasers to $AEVA (FMCW) and multiple unnamed CPO module makers; overlaps with $AXTI InP substrates.
- Packaging/test: Shunsin (讯芯) + MSSCorp (6830) sit downstream of both.
- Substrates: $SOI (SOI) and NGK (TFLN) are parallel material bottlenecks.
- Embedded substrates: NVDA/AMD/Intel interest flagged by @jukan05 as potential packaging shift away from traditional organic substrates.
- Geopolitical fork: Huawei hybrid-bonding route competes with the above Western/Taiwanese optical stack.
## 4. RISK & CATALYST WATCH
- Next week: OMX Stockholm inclusion + MSCI rebalance for $SIVE; $60 M+ passive inflow.
- Imminent: $SIVE NASDAQ listing (additional US institutional tranche).
- Fall 2026: Huawei Kirin chip launch using hybrid bonding (no EUV).
- Ongoing: AI-server MLCC lead-time extension (Nichidenbo commentary).
## 5. SIGNAL TABLE
| Ticker | Handle | Direction | Signal Type | Engagement | Thesis |
|---|---|---|---|---|---|
| AXTI | aleabitoreddit | bullish | early_trend | 2778 | 40 % InP chain control for $3–4 T AI capex |
| SIVE | aleabitoreddit | bullish | catalyst | 906 | OMX + MSCI + NASDAQ inflows next week |
| SIVE | aleabitoreddit | bullish | early_trend | 832 | First institutional passive wave into float |
| SOI | aleabitoreddit | bullish | early_trend | 1007 | Dominant SOI substrate monopoly |
| NVTS | aleabitoreddit | bullish | early_trend | 2444 | High-beta 10x power-semi name for 800 VDC |
| WOLF | aleabitoreddit | bullish | early_trend | 2444 | High-beta 10x power-semi name for 800 VDC |
| 6830 | aleabitoreddit | bullish | early_trend | 1278 | CPO yield-patent monopoly |
| 3363 | aleabitoreddit | bullish | confirmation | 385 | TSM COUPE optical module exposure |
No substantive investment posts from the other listed analysts in the last 24 hours.
MLCC & AI Power Delivery (Structural Shortage Thesis)
- @AntonLaVay: “为什么 MLCC 又重要了？” Full thesis: AI data centers shifting power architecture to 48V/54V and ±400VDC/800VDC (OCP Diablo, NVIDIA 800VDC AI factory); core problem is delivering hundreds-to-thousands of amps to <1V GPU/ASIC cores with fast load transients → voltage droop. High-end MLCC (low ESL, low height, high capacity, high reliability, land-side/die-side, server-grade) required for PDN impedance and decoupling; ordinary MLCC not in shortage. Structural, not cyclical: per-server MLCC count and specs rising; certification cycles long; Murata/Taiyo Yuden/Samsung Electro-Mechanics/TDK allocating capacity to high-margin specs. Silicon capacitors complement, not replace, at different frequencies. “大概率会” continue tight on high-capacity/small-size/low-ESL/high-voltage/high-temp specs. “MLCC正在从消费电子周期品的一部分，变成AI基础设施供电网络的一部分。” (includes images of A-share MLCC names; states “我没有深入了解每一家A股公司” and “推荐您自行研究”).
Macro / Recession Indicators
- @dampedspring: “Also January 2008” (reply context on chart/data point).
- @BickerinBrattle: “And that is called a recession” (reply context).
### Japan Value
-
@JapanDeepValue1: Quotes First Squawk on Japan’s financial regulator urging listed companies to channel more cash reserves into long-term business investment rather than buybacks and higher dividends. “I think speculation around this has been a huge part of it, it will be harder to ‘greemail’ these rock solid balance sheet companies so the activist trade is unwinding (to a degree)”
-
@JapanDeepValue1: Quotes own prior post on 6998 still trading below PBR 1x with underperforming chart vs other HDD-related names (reiteration of existing Japan value stance on the ticker).
### Investing Mindset / Conservatism (cross-domain for @ivanalog_com relative to primary tech/dividend focus)
- @ivanalog_com: “我和你一样，甚至更保守的多。我觉得每个人乐天知命就好。跟别人全仓，三倍比收益，我们显然是比不过的。别人的乐观和下注能力就决定了没法比。我有一天反思，一个是可能更接近物理真实的经历，另一个可能是2008年的A股给我的记忆太深刻了。每个人都有自己的福气和运气，不用太过自责。”
### Defense Tech (primary domain for @fundmyfund)
- @fundmyfund: Quotes Sam Rubin / Lockheed Martin NGSRI website post with GIF (no additional thesis or data stated).
### AI Infrastructure / Semiconductors
- @lithos_graphein: "Amazing security team. He's the trillion dollar man." (on NVIDIA CEO Jensen Huang spotted at Taipei’s Raohe St. Night Market)
### Substrates / Supply Chain (AT&S / ABF)
- @KairosPraxis: ""Delivering the product on time matters more than the exact cost" ~ Interesting insight from the CEO of AT&S on what their customers (chip manufactuers) actually care about. Context was whether they could pass on price increases (presumably in ABF raw material) to customers. Guess the real bottleneck is completing the shipment on time."
### AI Model Limitations
- @KairosPraxis: "This is the weirdest part to me. The models being barely aware of their own capabilities. I just had a frustrating conversation with Claude where it kept on hallucinating stuff about a document instead of telling me I forgot to attach it."
### SaaS Demand / AI Agents
- @OwlWealthy: "This feels like a 70% probability event in my mind - within the next 2 years demand for SaaS will be much higher due to the growth in the number of agents that will be using it and the volume of usage per agent Bullish for the right SaaS names $NOW $PLTR $MSFT $CRWD I also own a little $CRM"
### Semiconductor Process (Resist)
- @lithos_graphein: "For Broadband and i-line DNQ resist (not chemically amplified) you can tune the resist profile shape (taper/footing) with develop time and dose. The soft bake and PEB also affect it but then it's a full blown process DOE. FEM mandatory to get best focus for sure, assumed that was done. On a track the develop puddle can be limited over time so it's customary to do a second puddle to refresh it if you need longer develop time. Like 2x 60s puddle."
### Vendor vs Design Partner Dynamics
- @KairosPraxis: "Nice way to separate "design partners" from just vendors. If you just have a vendor, you can always dual-source."
### AI Infrastructure / Compute & Power Semi
- @TMTLongShort: "Supply of compute drives research org at labs which drives model capabilities which drives inference demand. Demand for intelligence is unlimited. The compute imbalance will only worsen until scaling laws break."
- @TheValueist: "$NVDA $MU $SNDK $LITE My Analyst Agent did some work for you @aleabitoreddit . I’ll update the analysis again as more responses come in. The power of GAI. POWER SEMI / 800 VDC CROWD-SOURCED PICK SCAN — UPDATED 05/24/26 12:40 PM ET Context: the tweet asked for highest-conviction “10x only” stock longs tied to the power semiconductor / 800 VDC AI data-center architecture theme, with NVTS and WOLF cited as examples. UPDATED PICKS BY OCCURRENCE • NVTS: 18 • WOLF: 12 • VICR: 11 • POWI: 8 • INFINEON (IFX / IFNNY): 6 • AOSL: 5 • FCEL: 5 • INNOSCIENCE / 02577: 5 • IPWR: 5 • OUST: 5 • HIVE: 4 • ATOM: 3 • CWR / CWR.L: 3 • SEDG: 3 • TE: 3 • BE: 2 • BRUN: 2 • DGXX: 2 • ENPH: 2 • EOSE: 2 • FLNC: 2 • HGRAF: 2 • IREN: 2 • KALRAY: 2 • LUMN: 2 • NBIS: 2 • NOK: 2 • RELL: 2 • VIVO: 2 • VLN: 2 Single-mention picks: ACLS, AEHR, AIP, AISP, AIXA, ALAB, ALKAL, AMBA, AMC, AMPG, AMSC, APH, ARBE, ASYS, CGEH, CPSH, CVV, DUOT, ENSI, EPOW, GFS, HYLN, IMSR, INDI, INFQ, INV, IONQ, IQE, KEEL, KQ / 168360, KULR, LAES, LPTH, MRVL, MSTR, NVEC, ON, PAYT, POW, PPSI, QS, RENESAS, RGTI, RKLB, SCE, SHLS, STM, TRT, TSLA, VIAV, VPG, VSH. LARGEST CHANGES VS PRIOR CAPTURE • NVTS: 10 → 18 (+8) • VICR: 6 → 11 (+5) • WOLF: 8 → 12 (+4) • Infineon: 3 → 6 (+3) • POWI: 6 → 8 (+2) • AOSL: 3 → 5 (+2) • OUST: 3 → 5 (+2) • New two-mention names: IREN, RELL Bottom line: new comments strengthened the same core cluster rather than changing the thread’s direction. NVTS remains the clear crowd favorite, WOLF remains #2, and VICR moved up materially. Infineon, AOSL, OUST, and Innoscience gained incremental support. RELL and IREN newly emerged as two-mention names."
Signal Book
| Ticker | Direction | Source | Thesis |
|---|---|---|---|
| VSH | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| VIAV | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| TSLA | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| STM | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| RKLB | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| QS | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| ON | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| MSTR | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| MRVL | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| IQE | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| IONQ | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| INDI | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| GFS | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| ENSI | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |
| APH | Long | @thevalueist | Single-mention 10x long tied to power semi/800VDC AI data-center architecture th… |