Pre-Market
Pre-Market Briefing — 2026-06-04
June 4, 2026Pre-Market Briefing — 2026-06-04
Market Regime
QQQ $746.16 — +13.2% vs SMA50 ($659.21) · Risk On ✅
Portfolio
| Strategy | Value | Return | DD | Positions | Cash |
|---|---|---|---|---|---|
| 4-Pool | $1,947,942 | +1847.9% | -2.0% | 20 | 0% |
| V5.3 | $1,268,945 | +1168.9% | 0.0% | 17 | 0% |
Breakout Signals
No breakout signals today.
_Section section_options_radar unavailable.
Open Positions
| Ticker | Entry | Current | Gain% | Strategy |
|---|---|---|---|---|
| 000660.KS | $611.02 | $1722.63 | +181.9% | 4-Pool |
| 0981.HK | $8.21 | $10.53 | +28.3% | 4-Pool |
| 1347.HK | $12.06 | $19.01 | +57.6% | 4-Pool |
| 2513.HK | $102.10 | $181.49 | +77.8% | 4-Pool |
| APLD | $32.19 | $47.86 | +48.7% | 4-Pool |
| APP | $433.51 | $605.63 | +39.7% | 4-Pool |
| AVGO | $402.17 | $481.57 | +19.7% | 4-Pool |
| CIEN | $67.36 | $627.00 | +830.8% | 4-Pool |
| CIFR | $15.42 | $26.29 | +70.5% | 4-Pool |
| GFS | $47.80 | $84.60 | +77.0% | 4-Pool |
| LITE | $62.86 | $1029.15 | +1537.2% | 4-Pool |
| LRCX | $246.49 | $334.41 | +35.7% | 4-Pool |
| MDB | $264.69 | $398.46 | +50.5% | 4-Pool |
| MP | $57.36 | $72.24 | +25.9% | 4-Pool |
| MU | $465.66 | $1064.10 | +128.5% | 4-Pool |
| NOK | $11.30 | $16.85 | +49.1% | 4-Pool |
| PWR | $317.65 | $706.06 | +122.3% | 4-Pool |
| TER | $107.65 | $392.62 | +264.7% | 4-Pool |
| TSM | $365.90 | $446.69 | +22.1% | 4-Pool |
| WULF | $18.05 | $26.49 | +46.8% | 4-Pool |
| 000660.KS | $611.02 | $1722.63 | +181.9% | V5.3 |
| 005930.KS | $138.98 | $263.14 | +89.3% | V5.3 |
| 0981.HK | $8.21 | $10.53 | +28.3% | V5.3 |
| 9888.HK | $15.68 | $17.31 | +10.4% | V5.3 |
| ASML | $1421.05 | $1705.37 | +20.0% | V5.3 |
| CIEN | $67.36 | $627.00 | +830.8% | V5.3 |
| CIFR | $15.42 | $26.29 | +70.5% | V5.3 |
| CRDO | $119.59 | $229.00 | +91.5% | V5.3 |
| ETN | $400.44 | $417.62 | +4.3% | V5.3 |
| FCX | $49.15 | $71.72 | +45.9% | V5.3 |
| GFS | $47.80 | $84.60 | +77.0% | V5.3 |
| GOOGL | $317.24 | $361.85 | +14.1% | V5.3 |
| LRCX | $246.49 | $334.41 | +35.7% | V5.3 |
| MDB | $264.69 | $398.46 | +50.5% | V5.3 |
| PWR | $317.65 | $706.06 | +122.3% | V5.3 |
| QCOM | $156.00 | $240.84 | +54.4% | V5.3 |
| TSM | $365.90 | $446.69 | +22.1% | V5.3 |
X Alpha — Analyst Chatter
Convergence Alerts — X + Breakout Scanner
| Ticker | Direction | Source | Breakout | Score | Thesis |
|---|---|---|---|---|---|
| ANET | Long | @thevalueist | VCP | 48 | Ethernet AI clustering beneficiary; high positive |
| ASX | Long | @thevalueist | Episodic | 48 | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advan… |
| FN | Long | @thevalueist | VCP | 45 | Optical components & 1.6T ecosystem exposure |
| ## 1. MACRO & THEMATIC OVERVIEW | |||||
| Hyperscaler AI capex has been revised sharply higher to a combined $5.3T (2025–2030) from $4.5T previously, with $GOOGL/$META/$AMZN cited directly. This spend is flowing into CPO and advanced packaging, where laser/photonics capacity and TSMC CoPoS are now visible bottlenecks. Two independent analysts converge on the same supply-chain pressure point: $SIVE positioned as the laser chokepoint across both NVDA NVLink and merchant/ASIC CPO routes, corroborated by $LITE’s EML allocation commentary. TSMC’s CEO simultaneously signals CoPoS pilot-line production already running with volume ramp expected over the next 2–3 years, reinforcing the packaging side of the same buildout. |
## 2. HIGH-CONVICTION IDEAS
$SIVE — Laser chokepoint inside every NVDA CPO and merchant/ASIC route
- Who: @aleabitoreddit (CPO/photonics supply-chain specialist)
- Thesis: $SIVE sits at the intersection of CW laser capacity constraints (signaled by $LITE earnings) and EML fab allocation; it appears in every NVDA NVLink CPO ecosystem partner plus merchant/ASIC CPO architectures. Technical dominance in these routes creates durable pricing power even if other pluggable players adopt novel 1.6T designs.
- Key data: Goldman hyperscaler capex $5.3T (2025–2030); CPO commercialization now viewed as H1–H2 2027 entry (earlier H2-2028 view revised forward).
- Valuation: Not quantified; thesis rests on capacity allocation and foundry (Win Semi) scale-up rather than multiples.
- Catalyst: Volume ramp at Win Semi private-placement capacity; design wins converting to production orders in 2027.
- Engagement: [825L 88RT 8QT] + [128L 15RT 0QT] + [98L 5RT 1QT] (multiple high-engagement threads).
- Contrarian?: Against short-seller narrative on insider sales; author explicitly bought the shares being sold.
- Cross-references: $LITE earnings cited as confirming laser bottleneck; no opposing view in batch.
TSMC — CoPoS pilot already running; >30% 2025 revenue growth locked in
- Who: @jukan05 (TSMC/ foundry specialist)
- Thesis: TSMC’s CoPoS advanced packaging is in pilot production with significant volume increase expected over the next two-to-three years; management simultaneously guides full-year revenue growth >30% driven by AI demand shifting from generative to agentic models.
- Key data: >30% FY revenue growth; CoPoS pilot already operating; Korea unable to replicate model in near term.
- Valuation: No multiples provided.
- Catalyst: CoPoS volume ramp 2026–2028; continued AI-driven foundry loading.
- Engagement: [349L 79RT 10QT] + [274L 70RT 1QT].
- Contrarian?: None in batch.
- Cross-references: Directly supports the CPO packaging side of the $SIVE thesis.
## 3. SUPPLY CHAIN & SECTOR MAP
$SIVE (laser/PIC) → CPO modules → NVDA NVLink ecosystem + merchant/ASIC CPO (Broadcom, Marvell, Astera, etc.).
CW laser capacity constrained; $LITE earnings indicate EML fab allocation crowding out other uses.
TSMC CoPoS advanced packaging sits downstream of the same CPO buildout; Win Semi private placement provides upstream foundry capacity for $SIVE. No direct overlap or conflict flagged between the two analysts.
## 4. RISK & CATALYST WATCH
- CPO volume timing: H1–H2 2027 entry (earlier H2-2028 view revised); low readiness still cited in some research reports.
- TSMC CoPoS volume ramp expected materially higher in 2–3 years.
- $SIVE insider-selling noise dismissed; CEO open-market purchase cited as counter-signal.
- Samsung/Meta custom SoC discussions suspended (Meta request) — negative read-through for non-TSMC ASIC routes.
## 5. SIGNAL TABLE
| Ticker | Handle | Direction | Signal Type | Engagement | Thesis |
|---|---|---|---|---|---|
| $SIVE | @aleabitoreddit | bullish | inflection | 1137 | Laser chokepoint across all NVDA + merchant CPO routes; $LITE EML allocation confirms bottleneck |
| $LITE | @aleabitoreddit | bullish | confirmation | 1025 | CW laser capacity already allocated to EML; signals broader photonics shortage |
| $GOOGL | @aleabitoreddit | bullish | catalyst | 1545 | Goldman raises combined hyperscaler capex to $5.3T (2025–2030) |
| $META | @aleabitoreddit | bullish | catalyst | 1545 | Goldman raises combined hyperscaler capex to $5.3T (2025–2030) |
| $AMZN | @aleabitoreddit | bullish | catalyst | 1545 | Goldman raises combined hyperscaler capex to $5.3T (2025–2030) |
| TSMC | @jukan05 | bullish | confirmation | 537 | CoPoS pilot running; >30% revenue growth guided; volume ramp 2–3 years |
No substantive investment posts found in the last 24 hours.
All retrieved posts from the specified handles (@_thevalueist, @antonlavay, @bearhunter, @bickerinbrattle, @bluechipdaily, @chartmaster, @concodanomics, @dampedspring, @dirtcheapstocks, @discountedtf) since 2026-06-03 were either replies without analysis, short non-thesis comments, or unrelated to markets/tickers (e.g., crypto option banter, Trump references, follower counts). No tickers, price targets, directional views, or theses emerged.
No substantive investment posts found in the last 24 hours.
The x_keyword_search returned only low-substance replies, GIFs, and personal comments (e.g., @fabknowledge on Knicks game energy; @fundmyfund short replies on $LPTH chart behavior, CEO comments, and drawdowns). None contained theses, data points, price targets, catalysts, or directional views.
No activity from the remaining handles. No themes emerged.
### AI Infrastructure (Google TPUv8t & Virgo Networking)
- @semianalysis_: "With the introduction of the TPUv8t, their new training focused TPU, Google unveiled a new scale-out network architecture called Virgo. Virgo is able to interconnect up to 134,400 chips with up to 47 Pbps of non-blocking bi-sectional bandwidth. (1/4)"
- @semianalysis_: "This marks a generational leap with up to 4x the bandwidth per accelerator and a 40% lower latency compared to previous generation. On a network perspective, this architecture connects 14 TPU pods (9,600 interconnected TPUs each, via 3D Torus ICI) on a flat 2-layers topology, with 14 planes of 150 switches per layer each. Google is using here Tomahawk 5 based high radix switches on the L1 layer and Tomahawk 4 switches on the L2 layer. (2/4)"
- @semianalysis_: "Within the plane, each switch is connected in full mesh using 800G 2xFR4 OSFPs in 8x100G breakout mode on the L1 layer and 400G FR4 QSFPs in 4x100G breakout mode on the L2 layer. To maintain a non-blocking topology, Google is doubling some links between the L1 and the L2 layers. Between the planes and the racks, 14 out of the 16 TPU trays within each rack connects to the corresponding plane (i.e. TPU tray 1 connects to plane 1, TPU tray 2 connects to plane 2, etc…), leaving 2 trays, or 4 NICs, to connect to the Jupiter Network. The NICs are directly connected to the plane using 400G FR4 QSFPs on the NIC side and 800G 2xFR4 OSFPs on the L1 switch side. (3/4)"
- @semianalysis_: "To learn more, please check our networking model: https://semianalysis.com/ai-networking-model/ (4/4)"
### Semiconductors Market Data & Dashboards
- @lithos_graphein: "Beta testing market data for my site now. This shows foreign markets too and will display each region's live trading. Need to watch it for a few days to iron out the bugs. Eventually I'm going to break these treemaps out for each tech industry page, and I'm even going to break semi down into Fab, Fabless, and Semitool. I also have plans to incorporate earnings alerts and other market info this dashboard. Lmk if you have suggestions or find this useful."
### M&A Activity (Tech / Semiconductors)
- @kairospraxis: "You've mentioned this before but the reason we're all so excited about the prospect of an m&a is these guys don't miss. All their acquisitions so far have been 🎯🎯 I'm usually not a fan of dilution but in this case I trust them."
- @owlwealthy: "$RBLX connecting the dots for their Reality vision" (quoting Roblox announcement of Morpheus AI acquisition)
### Optical Networking / DSPs
- @kairospraxis: "Okay that makes sense. Thanks. The interesting question is who is buying these 1.6 optical dsps? Credo doesn't have a 1.6T transceiver yet"
### AI Infrastructure Power, Electrical & Data Center Buildout
- @TheValueist: Broadcom call confirms AI gigawatt growth shifts bottleneck from chips to power/cooling/electrical infrastructure. Positive high-magnitude read-throughs for VRT, ETN, SU, GEV, ENR, PWR (near-term trading catalyst + multi-year shift). Broadcom’s AI XPV platform with Apollo/Blackstone targets >20GW compute capacity through 2028 ($35B 1st tranche); customers must align “power, the power shell.” Transmission: Vertiv (thermal/power distribution), Eaton/Schneider (switchgear/grid-edge), GE Vernova/Siemens Energy (generation/grid/transformers), Quanta (transmission/substations).
### US Power Generation & Capacity
- @TheValueist: Broadcom multi-gigawatt deployments (Anthropic, OpenAI, Meta, others) strengthen case for CEG, VST, NEE, NRG (medium-to-high magnitude positive, longer-duration fundamental shift). AI data centers need firm scalable power; supports nuclear/gas/renewables PPAs and capacity value in constrained markets.
### Alternative Asset Managers / AI Infrastructure Financing
- @TheValueist: Apollo/Blackstone gain new AI infra fee/deployment vector via $35B Apollo-launched tranche (medium magnitude positive near-term; risk-adjusted longer term). AI compute becoming private-capital asset class (equity, credit, leasing, fees).
### Enterprise Infrastructure, Servers & Private Cloud
- @TheValueist: Private cloud/enterprise AI inferencing accelerating. Positive for DELL, HPE, SMCI, 0992 (medium magnitude). Broadcom Q3 infrastructure software guide $8.9B (+31% YoY); VMware Cloud Foundation 9.1 driving on-prem deployments across AMD/Intel/NVIDIA.
### VMware vs. Displacement Narrative
- @TheValueist: Broadcom software momentum (ARR +17% YoY, Q3 guide +31% YoY) pressures VMware-displacement thesis. Negative for NTNX (medium magnitude).
### Hyperscaler Capex & FCF Pressure
- @TheValueist: >$30B Q2 AI semiconductor bookings, >$100B FY27 AI semi revenue reiterated, 2028 visibility. Negative medium magnitude near-term for GOOGL, META, MSFT, AMZN (elevated capex intensity despite custom silicon).
### Custom Silicon / Cloud Platforms (Google TPU, Frontier Labs)
- @TheValueist: Google long-term TPU agreement through 2031 materially de-risked (positive high magnitude for GOOGL). Custom XPU adoption creates longer-duration share pressure on NVDA/AMD merchant GPUs (negative medium magnitude longer term). Anthropic (>1GW 2026, additional 5GW 2027), OpenAI (10GW program, 1.3GW 2027 contractual), Meta (3GW through 2028, initial 1GW order 2H27).
### Ethernet AI Networking
- @TheValueist: Networking ~40% of Q2 AI revenue; positive high magnitude for ANET, medium for CSCO. Broadcom leadership in 100Tb/200Tb Ethernet switches, SerDes, Jericho fabrics. Ethernet gaining as default scale-out architecture.
### Optical Components & 1.6T Ecosystem
- @TheValueist: Positive medium-to-high for COHR, LITE, FN. Broadcom “de-facto standard” in CPO, 1.6Tb DSPs, CW/EML lasers; transition pulled by real AI cluster demand.
### HBM & Memory Supply Chain
- @TheValueist: HBM demand multi-year visibility (positive high for 000660, medium-to-high for MU/005930). XPU content per GW rising with multi-die/HBM; Broadcom “very comfortable” on 2026-2027 supply.
### Leading-Edge Foundry, Packaging & Test
- @TheValueist: Positive high for TSM; medium for ASE, BESI, 6857, TER, ASML, AMAT. >$30B Q2 bookings + FY26 $56B / FY27 >$100B AI semi guide tightens advanced capacity.
### Non-AI Semiconductors / Cyclical Recovery
- @TheValueist: Q2 non-AI semi revenue $4.2B (+6% YoY), bookings >$6B; Q3 guide $4.5B (+12% YoY). Positive low-to-medium for CSCO, WDC, STX, COMM (broadband/server storage/enterprise networking improving).
### Overall Cross-Portfolio Read-Throughs (Broadcom Q2 FY26 Call)
- @TheValueist: Highest-conviction positives: AI Ethernet/optical, HBM, leading-edge foundry/packaging, power/electrical infra, US power gen, AI data-center financing, private-cloud servers. Highest-conviction negatives: merchant GPU/proprietary networking share pressure, custom ASIC peer competition, hyperscaler capex/FCF burden, reduced VMware-displacement support. FY26 AI semi guide $56B; FY27 >$100B reiterated; visibility to 2028; 6 core customers. Q2 AI semi $10.8B (+143% YoY), bookings >$30B.
### Market Reaction / Portfolio Context
- @wey_how12640: June 3 cooldown on $SPX (retreated to 8ema); macro hawkish Fed/war concerns + $AVGO/$CRWD after-hours reactions weighing on hard/soft-ware. Portfolio holding up via core $SNDK/$MU (also green: $MRVL, $CIEN, $DOCN, $NOK); reversal in $NBIS/$BE. “Still early though.”
Signal Book
| Ticker | Direction | Source | Thesis |
|---|---|---|---|
| AVGO | Short | @wey_how12640 | After-hours reaction weighing on software names amid macro hawkish Fed/war conce… |
| COMM | Long | @thevalueist | Q2 non-AI semi revenue +6% YoY, Q3 guide +12% YoY shows cyclical recovery |
| STX | Long | @thevalueist | Q2 non-AI semi revenue +6% YoY, Q3 guide +12% YoY shows cyclical recovery |
| WDC | Long | @thevalueist | Q2 non-AI semi revenue +6% YoY, Q3 guide +12% YoY shows cyclical recovery |
| CSCO | Long | @thevalueist | Q2 non-AI semi revenue +6% YoY, Q3 guide +12% YoY shows cyclical recovery |
| AMAT | Long | @thevalueist | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advanced capacity |
| ASML | Long | @thevalueist | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advanced capacity |
| TER | Long | @thevalueist | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advanced capacity |
| ASX | Long | @thevalueist | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advanced capacity |
| TSM | Long | @thevalueist | >$30B Q2 bookings + FY27 >$100B AI semi guide tightens advanced capacity |
| MU | Long | @thevalueist | HBM demand multi-year visibility; XPU content per GW rising |
| FN | Long | @thevalueist | Broadcom de-facto standard in CPO/1.6T DSPs pulled by real AI cluster demand |
| LITE | Long | @thevalueist | Broadcom de-facto standard in CPO/1.6T DSPs pulled by real AI cluster demand |
| COHR | Long | @thevalueist | Broadcom de-facto standard in CPO/1.6T DSPs pulled by real AI cluster demand |
| CSCO | Long | @thevalueist | Networking ~40% of Q2 AI revenue; Ethernet gaining as default scale-out architec… |