Wilson

AI trends/tools • Market insights • Investing workflows
Charts/Insights are my opinion and NOT financial advice. Do you due diligence!

[Cryptic Trades @CrypticTrades_]

The S&P500 Seasonality over the past 30 years.

[Wilson @Inwaterfw]

$IREN 极具爆发力的追赶者

IREN并不是一个纯血的Neo Cloud, 不如NBIS那样有着完美的底层软件架构可以完美的支撑大模型推理应用。但作为一个半路转型的挑战者，我认为IREN可能有更强的后期爆发力.

IREN手握的电力资产是其最大的基本面护城河。IREN 目前在北美总计锁定了 约 2.9 GW的并网电力。这 2.9 GW是已经签署了实质性并网协议的容量。

具体到核心园区的分布和变现情况如下：

德州 Childress 园区（总计约 750 MW）：
现状： 大部分已通电（约 650 MW 处于运营状态）
核心 Deal： 1. 微软的百亿大单：微软此前已锁定该园区 200 MW 的专用容量，用于部署 Nvidia GB300 集群（一份长达多年、价值高达 97 亿美元的超级合同）。
2. 戴尔/英伟达订单：16 亿美元的风冷 Blackwell 系统，占用了该园区约 60 MW 的现有电力容量。

德州 Sweetwater 园区（旗舰级 2.0 GW 枢纽）：
Sweetwater 1 期 (1.4 GW）：这是 IREN 最大的王牌，直接光纤环网连接。该项目于 2026 年 4 月至 5 月进入通电阶段，是目前全美极少数能一次性提供GW级别增量电力的超大型 AI 数据中心。
Sweetwater 2 期 (600 MW)：于 2025 年初刚签署了 600 MW 的并网协议，支付了定金，排期在 2027 年底通电。

2. 潜在的电力储备与潜在 Deal
除了上述坐实的 2.9 GW，IREN 未来的催化剂主要集中在以下两方面的“潜在预期”：
在官方的开发者演示中，IREN 管理层明确表示，除了已经拿下的 2.9 GW，他们还有额外约 1.6 GW 至 2.0 GW 的项目处于“前期开发中”。这些属于尚未签署最终并网协议，但土地和初步电网评估已在推进的远期储备。

商业层面的潜在超级订单：极大概率在近期发生）：
巨大的“待租”电力敞口：虽然 IREN 有 2.9 GW 的电，但目前明确宣布被超大规模客户（如微软、戴尔）包下的算力消耗大约在 300 MW 左右。随着 Sweetwater 1 期（1.4 GW）的全面通电，IREN 手里即将拥有大量有电、有机房，等待 GPU 入驻的空白高密度算力位。

谁是潜在买家？ 彭博社等机构指出，美国目前有近 50% 的新建数据中心因为缺电而停滞。因此，IREN 剩余的这 1.0 GW+ 的现成电力，是所有大厂眼中的香饽饽。

潜在的 Deal 对象高度集中在：
1 其他 Hyperscalers：如急需算力的 Meta、AWS，Anthropic等极有可能包下 Sweetwater 动辄 100MW - 300MW 的单一厂房。
2 大型 GPU Neoclouds

与其他NeoCloud相比，IREN拥有电力，土地，以及配套的电力设施。而且已签署的极低的电价协议就是在这个AI缺电时代的最大的优势.

当然它的短板也非常明显:

1. 缺乏底层软件以及orchestration能力，管理层最近斥资 6.25 亿美元全资收购了全球领先的云基础设施和 Kubernetes编排专家 Mirantis就是在尝试补齐这块短板.
2. 半路出家缺乏信任，不过最近的Nvidia, Dell的背书也在逐渐的改善这点.
3. 从矿场到数据中心的机房改造是个吞金兽，即使最近超额发行了30亿的可转债，一样还有增发的风险。

$IREN

[Wilson @Inwaterfw]

$IREN 突破65，下一站70-75，逼空快要开始了。

ARR（年度经常性收入）指引从 37 亿美元上调至 44 亿美元。

与 $DELL 达成的价值 16 亿美元的合作协议

不要忘记了，5/7宣布的NVIDIA 的战略入股认购权：作为协议的一部分，NVIDIA 获得了一项为期5年的权利，可以以 $70 的行权价购买最多 3000 万股 IREN 股票。相当于英伟达的70$看涨期权，一旦突破75，英伟达一定会宣布认购…. 就会史诗级逼空. 下半年看100-110。

[Wilson @Inwaterfw]

🚀 $RKLB just proved it’s no longer “just a rocket company.”

This earnings report showed something much bigger:

Rocket Lab is becoming a vertically integrated space + defense + AI infrastructure company.

And Wall Street may still be underestimating it.

Here’s what stood out:

1️⃣ Revenue exploded to $200M
Up +63% YoY and above expectations.
Growth is accelerating — not slowing.

2️⃣ Backlog hit a record $2.2 BILLION
That’s future contracted revenue.

Aerospace companies live and die by backlog growth.
Rocket Lab’s keeps expanding.

3️⃣ They signed 36 new launch contracts in ONE quarter
More than all of 2025 already.

Demand for launch capacity is exploding.

4️⃣ Neutron changes everything
Most investors still don’t understand this.
Electron = small rocket business
Neutron = potential SpaceX competitor

That’s the real long-term unlock.

If Neutron succeeds:
Rocket Lab moves into:
• mega constellations
• defense launches
• larger payloads
• national security missions
That massively expands TAM.

5️⃣ Defense exposure is ramping hard
Rocket Lab signed:
• hypersonic contracts
• Pentagon programs
• missile defense partnerships
This isn’t just “space hype” anymore.

It’s becoming defense infrastructure.

6️⃣ The company is vertically integrating aggressively
They acquired:
• satellite communications tech
• robotics systems
• optical networking
This matters.
The winners in the new space economy likely control:
launch + satellites + software + communications.
Rocket Lab is building the full stack.
7️⃣ AI is quietly part of the story

Most people miss this connection.

AI needs:
• satellites
• defense systems
• edge computing
• autonomous infrastructure
• global communications
Space infrastructure becomes more valuable in an AI-driven world.
That’s why RKLB is increasingly being discussed alongside: $PLTR $LMT $NOC and even SpaceX.

The biggest takeaway:
Rocket Lab is evolving from:
“small satellite launcher”

→ into a full aerospace + defense platform.

And if Neutron executes successfully over the next few years…

this company could enter an entirely different tier.

The space economy is still early.
And RKLB is positioning itself at the center of it. 🚀

[Wilson @Inwaterfw]

Most investors have never heard of $MRAM.
MRAM broke above major resistance and 52-week highs
Trading volume surged far above normal levels

Everspin has recently gained attention for:
• defense contracts,
• aerospace applications,
• radiation-hardened memory,
• and persistent solutions for AI/edge systems.

Its technology could become critical for the AI era.
Why?

Because AI has a memory problem.

Current memory systems are:
• too power hungry
• volatile
• vulnerable to data loss

That’s where MRAM comes in.
Everspin Technologies is building STT-MRAM:
(Spin-Transfer Torque Magnetoresistive RAM)
A next-gen memory technology combining:
⚡ DRAM-like speed
💾 Flash-like persistence
🔋 Low power usage
🛡️ Extreme durability

Think of it as:
“persistent high-speed memory.”

Here’s why that matters for AI:

Robots/autonomous systems need instant-on memory
Edge AI devices need durable low-power storage
Defense + aerospace require radiation-resistant memory
MRAM stores data magnetically instead of electrically.

Meaning:
• no constant refresh cycles
• near-instant recovery after power loss
• extremely high endurance
• lower latency in critical workloads

Potential AI-era use cases:
→ AI servers & accelerators
→ robotics
→ autonomous vehicles
→ drones
→ satellites
→ defense systems
→ industrial AI
→ edge computing

The most interesting part:
As AI moves from “chatbots” → “physical AI,”
memory reliability becomes massively important.
That could make MRAM one of the hidden infrastructure layers of the AI stack.
The AI race isn’t just about GPUs anymore.
It’s becoming:
compute + power + networking + memory architecture.

Most people only see the surface layer.

The real money is often hidden deeper in the stack.
Key Technology Summary (for carousel/thread continuation)
What MRAM Actually Is
MRAM = Magnetoresistive Random Access Memory
Unlike DRAM or NAND flash:

it stores data using magnetic states

not electrical charge
That gives it:
persistence
lower standby power
high endurance
instant recovery after power failure

Researchers are actively exploring STT-MRAM for AI accelerators and ML systems.

Where MRAM Is Used

Current / Emerging Applications

Data Centers
Persistent write buffers.

Aerospace & Defense

Radiation-resistant persistent memory for:
satellites
avionics
drones
Industrial AI

Real-time systems needing:
instant-on operation
deterministic performance
Robotics & Autonomous Systems
Persistent AI state storage with low power requirements.

[Wilson @Inwaterfw]

Most people still think AI is “just ChatGPT.”

They’re missing the bigger picture.

We’re only in Year 4 of a 15+ year AI supercycle.

The pattern looks like this:

Phase 1 → Build the chips
Phase 2 → Scale the infrastructure
Phase 3 → Physical AI economy
Phase 4 → AI dominates software & enterprise
Here’s how I see it unfolding:
🟣 PHASE 1 (2023–2025) — FOUNDATION
The winners were:
$NVDA $AMD $MU $AVGO $TSM
This phase was about:
• GPUs
• memory
• semiconductors
• AI training infrastructure

Most of this move already happened.

🔵 PHASE 2 (2025–2027) — INFRASTRUCTURE EXPANSION
This is where we are now.
The bottleneck shifts to:
• data centers
• networking
• power
• cooling
• electricity
Stocks benefiting:
$VRT $ETN $ANET $MRVL $CEG $IREN $LITE

AI needs massive infrastructure to scale.
🟢 PHASE 3 (2027–2029) — THE PHYSICAL AI ECONOMY
This is where AI leaves the screen and enters the real world.
Think:
• robotics
• autonomous systems
• logistics
• defense
• drones
• industrial automation
Potential leaders:
$TSLA $RKLB $ASTS $SYM $KTOS $MP
This phase could be bigger than most expect.

🟠 PHASE 4 (2030+) — AGI & SOFTWARE DOMINANCE
Once infrastructure is mature:
AI becomes embedded into everything.

Likely leaders:
$MSFT $GOOGL $AMZN $META $ORCL
And eventually:
Quantum + AGI infrastructure.

The biggest mistake investors make is thinking AI is a 1-year trade.
This is likely a decade-long transformation cycle.
Infrastructure → Expansion → Domination.
We are still early.

[Wilson @Inwaterfw]

@StockOptions888 $SPX

[Wilson @Inwaterfw]

@LuxAlgo Thanks! What indicator on the chart?

[Wilson @Inwaterfw]

12|7|24: $COIN EOW update:
- Strong weekly breakout and overall relative strength compared to other crypto stocks.
- 355 - 370 is the last resistance before hitting IPO high.
- Large weekly bull flag measured move target 430-440.

https://t.co/WiFIlMDwbR

[Wilson @Inwaterfw]

11|11|24: $COIN update:
- Keep flying with $BTC!
- We gapped up above 300, and a daily close above 300 would be super bullish to seal the path to 370, 430!

https://t.co/eXBUU7uBuG

[Wilson @Inwaterfw]

12|7|24: $AUR New idea:
- Strong weekly breakout of the 2+ years round bottom.
- w/ more friendly regulation of autonomous driving expected, this name might have a big upside room.
- Short term target 9.5, then 11.5-13. Avoid/SL below 6.5.

[Wilson @Inwaterfw]

12|7|24: $RXRX New idea:
- Another $NVDA invested AI bio stock which is still at its bottom.
- Increased weekly volume, Weekly MACD golden cross.
- 15 - 16 not a bad target.

[Wilson @Inwaterfw]

12|6|24: $AI EOW update:
- Gave you the chart idea @ 27 several weeks ago and it's now up 50%!
- ER next Monday which is almost binary event. As long as 33 holds post ER, this is going to 45-50 next!

https://t.co/Sgor7dIz1p

[Wilson @Inwaterfw]

10|13|24: New idea $AI
- Beautiful weekly chart w/ breakout imminent.
- Nice consolidation in last 3 months. Weekly MACD bullish divergence and golden cross last week.
- Tight SL ~23.7, good R&R to play upside.

[Wilson @Inwaterfw]

12|6|24: $DOCU update:
- What a massive move post ER, all targets met!

https://t.co/y6BjU3BjcG

[Wilson @Inwaterfw]

11|9|24: $DOCU Path to 88-92, then breakout measured move ~100.

[Wilson @Inwaterfw]

12|4|24: $HIMS update:
- Every dip gets bought! Once it breaks above 35, we shall see 42 next and my mid-term target is ~50.

https://t.co/ye5Bzzju7V

[Wilson @Inwaterfw]

11|11|24: $HIMS update
- Gapped up above 24 and now is +19%, did you catch it? 😀

https://t.co/mzIEMlWSMZ

[Wilson @Inwaterfw]

12|4|24: $HOOD update:
It is unstoppable, we are going to hit 44 - 45 by the end of this month!

https://t.co/LzVqboHAGE

[Wilson @Inwaterfw]

11|11|24: $HOOD update
- Hit my 33-34 target early! Took profits and rolled the long positions to later expiration.

https://t.co/irbqVF2D6z

[Wilson @Inwaterfw]

@htnhufi1 互关

[Wilson @Inwaterfw]

11|23|24: $AI EOW update:
- Finally breakout the 3 months consolidation with strong volume that I have been waiting for!
- As long as 33 level holds, it should move towards 45-50 weekly resistance zone. A lot of room above!

https://t.co/Sgor7dIz1p

[Wilson @Inwaterfw]

10|13|24: New idea $AI
- Beautiful weekly chart w/ breakout imminent.
- Nice consolidation in last 3 months. Weekly MACD bullish divergence and golden cross last week.
- Tight SL ~23.7, good R&R to play upside.

[Wilson @Inwaterfw]

11|23|24: $DDOG EOW update:
- God candle last week with 22% up!
- As projected, it hit my 153-155 target in Nov!
- Currently in a strong weekly resistance zone, if it can breaks above, it could hit 190 - 200 in Q1 next year.

https://t.co/9v7TyjyliF

[Wilson @Inwaterfw]

10|12|24: $DDOG weekly update:
- Nice run to 131.2 after my chart idea of 122.7. 1st target hit.
- From weekly TF, it still has a lot of steam to run, increased weekly volume and this is just the first week of MACD golden cross.
- If it can break above 135 in next week or two, I think it can run to 153-155 in Nov/Dec.

https://t.co/zdfVPN1QaA